MXPA97002273A - Method and device for demodulation and detection of power control bits in an extend spectrum communications system - Google Patents

Abstract

The present invention relates to a receiver circuit (400, 500) receiving an extended spectrum communications signal, for example a DS-CDMA signal, which includes a pilot channel and a power control allocator. The signal is reduced and decoded. The pilot symbols in the pilot channel are provided to a channel estimator (408) for estimating the channel phase and the channel gain of the communications channel. This estimate is provided to a demodulator (422) to demodulate the traffic channel symbols. The pilot symbols are provided to another channel estimator (410) to estimate the channel phase and the channel gain for the power control allocator. This estimate is provided to a demodulator (424) to demodulate the power control allocator. The traffic channel symbols are delayed by a predetermined time in a delay element (420) before their demodulation. The power control allocator is delayed a short time or is not delayed at all in a short delay element (418) before its demodulation.

Description

METHOD AND DEVICE FOR DEMODULATION AND DETECTION OF POWER CONTROL BITS IN A COMMUNICATIONS SYSTEM EXTENDED SPECTRUM Cross Reference to Related Requests This request is related to the application form number 08 / ****** (file number CE01019R), entitled "Method and Device for Demodulation for Extended Spectrum Communication with a Pilot Channel", presented on the same date as the present one and assigned to the assignee of this invention.

Field of the Invention This invention relates to extended spectrum radio communications. This invention relates more particularly to a method and apparatus for the demodulation and detection of power control bits for use in a receiver in an extended spectrum communication system using a pilot channel.

Background of the Invention Radio systems provide wireless communications to users of radio subscriber units. A particular type of radio system is a cellular radiotelephone system. A particular type of radio subscriber unit is a cellular radiotelephone subscriber unit, sometimes referred to as a mobile station. Cellular radiotelephone systems generally include a switch controller connected to the public switched telephone network (PSTN) and a number of base stations. Each of the number of base stations generally defines a geographic region close to the base station to produce coverage areas. One or more base stations communicate with a base station that coordinates a call between the mobile station and the public switched telephone network. The base station offers radiotelephone communication service between mobile stations operating in a cell and the public switched telephone network (PSTN). The communication connection for a carrier signal from the base station to the mobile station is referred to herein as a downlink connection. Conversely, the communication connection of a mobile station to the base station is referred to herein as an uplink. A description of a cellular radiotelephone system is available in the book "Cellular Mobile Communications Systems" by William C.Y.Lee, 1989.

A particular type of cellular radiotelephone system employs emission of extended spectrum signals. The emission of extended spectrum signals can be broadly defined as a mechanism whereby the bandwidth occupied by a transmitted signal is much wider than the bandwidth required by a baseband information signal. The two categories of extended spectrum communications are Direct Sequenced Extended Spectrum (DSSS) and Extended Frequency Hopping Spectrum (FHSS). The spectrum of a signal can be extended more easily by multiplying it with a signal generated by broadband pseudorandom code. It is essential that the extension signal is precisely known so that the receiver can reduce the signal. The essence of the two techniques is to extend the power transmitted from each user in such a wide bandwidth (1-50 MHz) that the power per unit bandwidth, in watts per hertz, is very small.

The emission of extended spectrum signals provides improved performance with respect to narrow band techniques. The systems of frequency jump reach their gain of processing preventing interferences. Direct sequence systems use an interference attenuation technique. For the DSSS, the objective of the receiver is to choose the signal transmitted from a received wide bandwidth in which the signal is below the background noise level. To do so, the receiver must know the carrier signal frequency, the modulation type, the pseudorandom noise code relation and the code phase, since the signal-to-noise ratios are generally -15 to 30 dB. The determination of the phase of the code is the most difficult of these.

The DSSS technique acquires superior noise performance, compared to the frequency jump, at the expense of increasing the complexity of the system. In addition, the DSSS receiver must lock and track the correct phase of the received signal within a time in chips (ie a partial bit period).

A cellular radiotelephone system using DSSS is commonly known as a Direct Sequence Codes Division Multiple Access (DS-CDMA) system, in accordance with the IS-95 standard of TIA / EIA. Each user in the system uses the same radiofrequency frequency but is separated from other users by the use of a particular extension code. Other extended spectrum systems include radiotelephone systems operating at 1900 MHz, commonly referred to as DCS1900. Other radio and radiotelephone systems also use extended spectrum techniques.

In an extended spectrum communication system, downlink transmissions include a pilot channel and a number of traffic channels. The pilot channel is decoded by the users. Each traffic channel is intended to be decoded by a single user. Accordingly, each traffic channel is coded using a code known to both the base station and the mobile station. The pilot channel is coded using a code known both by the base station and by all mobile stations.

The pilot channel serves many purposes. Among these we find timing and carrier phase synchronization in the receiver of a base station, estimation of the channel gain and the phase shift imposed by the channel, for the diversity combination and the convolutional soft decoding. The performance of the mobile station receiver depends on the accuracy of the channel phase estimation and the channel gain. In the receiver, the pilot channel signal is reduced to obtain a reduced channel signal. The reduced pilot channel signal contains channel information, including channel phase and channel gain, which is corrupted by noise and interference. More accurate channel phase and gain information must be extracted from the reduced pilot channel signal for demodulation and decoding.

Conventionally, estimates of the channel phase have been generated separately from channel gain estimates. In general, the phase of the reduced pilot channel signal has been used to lead to a locked phase loop that generated a more accurate channel phase estimate to be used for coherent demodulation. The magnitudes of the reduced pilot channel symbols, or their squares, were averaged to generate a channel gain estimate when this amount was needed, for example for diversity combination and soft decoding.

While this implementation using a locked phase loop can provide adequate performance in many situations, performance may be limited when the quality of the communication channel is marginal. In such situations, a better method and apparatus for the demodulation of the extended spectrum communication signal is necessary.

In addition to the normal pilot channel and traffic channel signals, the downlink transmissions also include a power control indicator in the traffic channel. The power control indicator is transmitted by the remote base station to the mobile station to control the transmit power of the mobile station. The power control indicator conventionally includes several bits that are not encoded in any way. In response to the power control indicator, the mobile station adjusts its transmission power to adapt it to the caiabant channel conditions, for example fading or blocking or the sudden absence of these. To achieve reliable and accurate communication, a rapid response from the mobile station to the received power control indicator is necessary.

Accordingly, a method and apparatus for the demodulation of an extended spectrum communication signal, including accurate and rapid detection of the power control bits, is needed in the art.

Brief Description of the Figures The features of this invention, which are considered new, are set forth in particular in the appended claims. The invention, together with other objects and advantages thereof, can be better understood by reference to the following description, taken together with the accompanying drawings, in which figures the same reference numbers identify identical elements and where: Fig. is an operating block diagram of a mobile radiotelephone station. Fig. 2 is a block diagram of a first filter for use in the mobile radiotelephone station of Fig.l. Fig.3 is a block diagram of a second filter for use in the mobile radiotelephone station of Fig.l. Fig. 4 is an alternative block diagram of a receiver circuit for use in the mobile radiotelephone station of Fig.l. Fig. 5 is a second block diagram of alternative operation of a receiver circuit for use in the mobile radiotelephone station of Fig.l. Fig.6 is a block diagram of operation of a power control channel estimator for use in the receiver circuit of Fig.4.

Detailed Description of the Preferred Embodiment With reference to FIG. 1, a block diagram of operation of a mobile radiotelephone station 100 is shown. The mobile station 100 includes a first antenna 102, a second antenna 104, a first filter circuit 106, a second filter circuit 108, an antenna switch 110, a first receiver indicator 112, a second receiver indicator 114, a third receiver indicator 116, a combiner 118, a decoder 120, a controller 122, a subscriber interconnection 124 , a transmitter 126 and an antenna switch 128. The mobile station 100 is preferably configured for use in a DS-CDMA cellular radiotelephone system that includes a number of remote location base stations. Each base station includes a transceiver that emits and receives radio frequency signals (RF) to and from the mobile stations, including the mobile station 100, within a fixed geographic area. This is an application for the mobile station 100, but the mobile station 100 can be used in any suitable extended spectrum communication system.

In the mobile station 100, the first antenna 102 and the second antenna 104 emit and receive radio frequency signals to and from a base station (not shown). The radiofrequency signals received in the first antenna 102 are filtered, converted from analog signals to digital data and processed in the first filter circuit 106. Similarly, the radio frequency signals received in the second antenna 104 are filtered, they convert analog signals into digital data and are processed in the second filter circuit 108. The first filter circuit 106 and the second filter circuit 108 can also fulfill other functions such as automatic gain control and intermediate frequency downconversion ( IF) for processing.

In an alternative embodiment, the mobile station 100 may include only a single antenna and a single filter circuit. However, the provision of two antennas and filter circuits associated with them provides space diversity to the mobile station 100. In a system in diversity of space, a signal travels slightly different paths of the transmitter to the two antennas in the receiver, due to multipath reflection or other reasons. Although the trajectory of the transmitter to one of the two antennas may produce phase cancellation of the transmitted and reflected path waves, it is less likely that multiple paths to the other antenna will simultaneously cause phase cancellation. The antenna switch 110 selects between the first antenna 102 and the second antenna 104 as the source of received radiofrequency signals.

The mobile station 100 preferably employs a tilt receiver that includes the first receiver indicator 112, the second receiver indicator 114 and the third receiver indicator 116 for receiving a spread spectrum communication signal over a communication channel. The tilt receiver design using multiple indicators is conventional. The output signals of each indicator of the tilt receiver are combined by the combiner 118. The structure and operation of the first receiver indicator 112 will be discussed in more detail below. Preferably, the second receiver indicator 114 and the third receiver indicator 116 operate essentially the same as the first receiver indicator 112.

As noted, the combiner 118 combines the output signals of the tilt receiver indicators and forms a received signal. The combiner 118 supplies the received signal to the decoder 120. The decoder 120 may be a Viterbi decoder or another type of convolutional decoder or other suitable decoder. The decoder 120 retrieves the data transmitted on the radio frequency signals and sends the data to the controller 122. The controller 122 formats the voice data or information recognizable for use by the subscriber interconnect 124. The controller 122 is connected electrically with other elements of the mobile station 100 to receive control information and provide control signals. The control connections are not shown in Fig. 1 so as not to unduly complicate the drawing. The controller 122 generally includes a microprocessor and a memory. The subscriber interconnection 124 transmits the information or voice received to the subscriber. In general, the subscriber interconnection 124 includes a screen, a keyboard, a speaker and microphone.

After the transmission of radio frequency signals from the mobile station 100 to the remote base station, the subscriber interconnect 124 transmits input data to the controller 122. The controller 122 formats the information obtained from the subscriber interconnect 124 and the communicates to transmitter 126 for conversion into modulated radiofrequency signals. The transmitter 126 communicates the modulated radiofrequency signals to the antenna switch 128. The antenna switch 128 selects between the first antenna 102 and the second antenna 104 for transmission to the base station.

The structure and operation of each of the tilt receiver indicators to receive and demodulate the signals is described below, using the first receiver indicator 112 as an example. In accordance with this invention, the mobile station 100 is configured to receive an extended spectrum communication signal, preferably a direct sequence code division multiple access signal (DS-CDMA), over a communication channel. The extended spectrum communication signal includes a pilot channel and a number of traffic channels. In a transmitter, for example the base station in a cellular radiotelephone system, the pilot channel and the traffic channels are coded using different Walsh codes. In general, the pilot channel is coded using a Wals (0) code, a first traffic channel is coded using a Walsh code (2), etc. After coding, the spectrum of the signal is extended using a pseudorandom (PN) noise code. The spread spectrum signal in digital form comprises a series of chips whose respective values are defined by the PN code and the encoded data. The Walsh coding for each traffic channel is unique to that channel and the recipient receiver. Each receiver in the system, or subscriber in the cellular radiotelephone system, is assigned a unique Walsh code corresponding to the traffic channel in which it communicates with the base station to decode the traffic channel. Each receiver also decodes the pilot channel. In accordance with this invention, the pilot channel is used to estimate the channel phase and channel gain of the communications channel.

The first receiver indicator 112 includes a reducer 150, a decoder 151, a pilot channel adder 152, a filter 154, a conjugate generator 156, a traffic channel decoder 158, a traffic channel adder 160, an element in delay 162 and a demodulator 164. Those skilled in the art will recognize that these elements can be implemented in hardware or software, or in some combination of both that increases manufacturing efficiency and feasibility.

The reducer 150 receives from the antenna switch 110 a digital representation of the extended spectrum communications signal received by the mobile station 100. The reducer applies a pseudorandom (PN) noise code to the received signal. The reducer reduces the received signal, producing a reduced signal. The PN code is stored in the mobile station 100 and can be transmitted to the mobile station 100, for example from a base station, when the communication channel between the base station and the mobile station 100 is started. The PN code it is unique to the mobile station 100, so that no other receiver in communication with the base station can decode the traffic channel transmitted to the mobile station 100.

The reduced signal is provided by the reducer 150 to the pilot channel decoder 151. The pilot channel decoder applies a pilot channel code to the reduced signal to produce the pilot channel signal. The pilot channel code is usually the Wals code (0). The pilot channel docoder applies the decoded signal to the pilot channel adder 152. The pilot channel adder 152 includes an adder 166 and a switch 168. The adder 166 adds 64 consecutive chips to form a pilot symbol. After each chip number sixty-four, the switch 168 closes to connect the adder 166 with the filter 154 to provide a pilot symbol received to the filter 154. Therefore, the channel adder 152 detects the pilot channel.

The embodiment shown in Fig. 1 is suitable if a Walsh code is used to encode the pilot channel. Since Wals (0) includes all ones, no decoding is needed when the pilot channel is encoded using Walsh (0) and the pilot channel decoder can be omitted. However, if another Walsh code or other coding is used to encode the pilot channel, a decoder is necessary. That decoder applies a pilot code to the reduced signal to produce the pilot channel signal. In the preferred embodiment, the pilot code is common to all subscribers in communication with the base station.

The filter 154 receives the pilot symbols of the pilot channel adder 152. The filter 154 filters the pilot channel signal to obtain a complex representation of an estimated channel gain and an estimated channel phase for the communication channel, in a form that is described below. it is known from the theory that, if the real channel gain lh (n) l and phase fh (n) at time nT are known, the optimal demodulation can be implemented according to: (1) e'j? h < n > r (n) where r (n) is the traffic channel symbol at the output of the traffic channel adder 160. The optimal weighted value (maximum likelihood) used in the combination is the real part of for (coded) bits in a BPSK modulated symbol in nT and the real and imaginary parts of (2) for the two bits in a QPSK modulated symbol in nT, respectively, as long as the noise is stable and has the same variation for each indicator or antenna of the mobile station 100 The quantity given by (2) can be rewritten as: (3) h * (n) r (n) where is the complex representation of the channel coefficient. For a mobile fading channel, h (n) is a random low pass process. The highest frequency in the spectrum of h (n) is equal to the Doppler frequency for a mobile communication channel.

Since the complex channel coefficient is not known, it is necessary to estimate the magnitude and phase of the channel coefficient. The estimated channel coefficient is used instead of its real value for the demodulation and generation of weighted values in the receiver. Specifically, denoting h (n) is an estimate of h (n), the value of weigher to combine and decode is computed as the real and imaginary parts of A (5) h * (n) r (n). It is possible to estimate the phase and channel gain together using the pilot symbol.

The pilot symbol can be expressed as: (6) p (n) = a [h (n) + z (n)] where a is a constant that depends on the implementation of the receiver and z (n) stable additive or interference white noise. Since it does not change once the receiver has been designed, without loss of generality, we leave CF = 1.

The pilot symbol p (n) can be used as an estimate of h (n). However, a more accurate estimate of h (n) can be obtained by averaging some p (n), so that K2 (7) h (n) =? w (k) p (n-k), k = -K! where w (k) are the weighted coefficients. When K_ > 0, the delay must be entered before the demodulation can be carried out.

The optimal weighted coefficients w (k) can be computed (8) = R "1 F where the vector = [w (-K?), ..., w (0), ..., vj (K2)] t , R is the autocorrelation matrices of p (nk) and F is the cross correlation vector between p (nk) and h (n) These values can be computed if the statistic of h (n) is known.

When the statistics of the channel variation is known, the optimal weighted coefficients can not be determined accurately. An example of this situation occurs when the Doppler frequency changes during a communication session and the receiver only knows the maximum value of the Doppler frequency. In this case, the weighing coefficients have a low-pass frequency response. The maximum Doppler frequency of the channel should be within the passband of this low pass response.

The filter 154 is preferably a low pass filter. The filter input is the pilot symbol p (n). The output of the filter is the estimate? H (n) of the channel coefficient. ? h (n) is a complex number that contains phase and magnitude information. The phase information corresponds to a channel phase estimate. The magnitude information corresponds to a channel gain estimate. Possible implementations of the filter 154 will be described below in conjunction with Figs.2 and 3. The conjugate generator 156 determines the complex conjugate of the signal h (n) produced by the filter 154. The filter 154, in conjunction with the generator of conjugate 154, produces an estimate of the complex conjugate of the complex representation of the channel gain and the channel phase for the communication channel. The complex conjugate of the complex representation of the channel phase and the channel gain is provided to the demodulator 164.

The reduced signal is also provided from the reducer 150 to the decoder 158. The decoder 158 applies a specific subscriber traffic code to the reduced signal to produce the traffic channel signal. The subscriber-specific traffic code is the Walsh code (n) assigned to the mobile station 100. The traffic channel signal is provided to the traffic channel adder 160.

The traffic channel adder 160 includes an adder 170 and a switch 172. The adder 170 adds 64 consecutive chips to form a traffic symbol. After each chip number sixty-four, the switch 172 closes to connect the adder 170 to the delay element 162 to provide a received traffic symbol to the delay element 162. Therefore the traffic channel adder 160 detects the channel of traffic. More specifically, the traffic channel adder 160 detects the traffic symbol r (n).

The delay element 162 is preferably a FIFO, or regulator first enters, first exits. The filter 154 introduces a flip delay when the channel gain and channel phase are estimated. The delay element 162 compensates for the filter delay to ensure that the estimated channel phase and estimated channel gain are used to demodulate the corresponding traffic symbols. The delay element 162 delays the spread spectrum communication signal a predetermined time to produce a delayed signal. More specifically, the delay element 162 delays only the traffic symbols of the traffic channel to produce the delayed traffic symbols.

The inventors have determined that the delay of the traffic symbols by 0.5 to 2 milliseconds gives the best results in a DS-CDMA cellular radiotelephone. More specifically, the inventors have determined that a delay of 31 symbols, corresponding to 1.5 milliseconds, produces the best results. The performance of the receiver under these conditions is only 0.15 dB of the performance of the ideal receiver (unattainable) that uses channel gain and channel phase. However, reducing the delay to 0.5 milliseconds and using an appropriate filter will give little degradation in receiver performance. - The delayed traffic symbols are provided to the demodulator 164. The demodulator 164 can be implemented as a multiplier which multiplies the delayed traffic symbols and the signal received from the conjugate generator 156, demodulating the delayed traffic symbols using the estimated channel phase and the estimated channel gain. The result of this multiplication is provided to the decoder 120 for further processing.

Referring now to FIG. 2, a block diagram of a finite impulse response (FIR) filter 200 for use in the mobile radiotelephone station 100 of FIG. 1 is shown therein. The filter 200 can be used to provide the low pass filter function of the filter 154 in Fig.l. The filter 200 includes delay elements 202,204,206, multipliers 208,210,212,214 and an adder Preferably the filter 200 uses a total of 61 delay elements such as delay elements 202,204,206, not all of which are shown in Fig. 2 so as not to unduly complicate the drawing. The delay elements operate in phases in sequence, changing the pilot symbols in series through the chain of delay elements. The delay elements are connected in series so that, during a first phase, the delay element 202 receives a first pilot symbol from the pilot channel adder 152 (Fig. 1). After a delay equal to a pilot symbol period, during a second phase, the first pilot symbol is transmitted from the delay element 202 to the delay element 204 and a second pilot symbol is transmitted from the pilot channel adder 152 to the element delay 202. Again, after a delay equal to a pilot symbol period, during a third phase, the first pilot symbol is transmitted from the delay element 204 to the next delay element connected in series with the delay element 204, the second pilot symbol is transmitted from the delay element 202 to the delay element 204, and a third pilot symbol is transmitted from the pilot channel adder 152 to the delay element 202.

During each phase, the pilot symbols stored in each delay element are multiplied with a weighing coefficient by a respective multiplier 208,210,212,214. Preferably the filter 200 uses a total of 62 multipliers such as multipliers 208, 212, 212, 14, not all of which are shown in Fig. 2. Each multiplier corresponds to one of the delay elements 202,204,206. The multipliers multiply the delayed pilot symbol stored in the respective delay element by a weighing coefficient. Likewise, the multiplier 208 multiplies the incoming pilot symbol, at the input of the delay element 202, by a weighing coefficient.

The weighing coefficients w (k) are preferably calculated according to the preceding equation (8). Alternatively, the weighting coefficients can be estimated according to any appropriate method. In a simple example, all the weighing coefficients w (k) can be set equal to unity. In such an implementation, the filter 200 is a low pass filter that averages a predetermined number (eg 42) of pilot symbols without weighing. Preferably, the weighing coefficients w (k) are chosen such that the filter 200 has a frequency response close to the low pass response described above. Therefore, the filter 200 operates to take samples of a predetermined number (for example 61) of pilot symbols, multiply the pilot symbols sampled by weighted coefficients and combine the products to produce a complex representation of the channel gain estimate and channel phase.

In an alternative embodiment, the filter 154 (Fig. 1) could be implemented using a low-pass infinite impulse response (IIR) filter. This IIR filter should have a quasi-linear phase response within this bandpass.

The filter 154 is characterized by a group delay in the frequency of interest. For a linear phase FIR filter, for example filter 200, the group delay of the filter is equal to half the delay or length of the filter.

For a FIR filter or a non-linear phase IIR filter, the group delay is defined as df (f) df | MO | where f is the phase rotation introduced by the filter at the frequency f and fo is the frequency that interests us. In accordance with this invention, the delay introduced by the delay element 162 is essentially equal to the group delay of the filter 154.

Fig. 3 is a block diagram of a filter 300 for use in the radiotelephone base station of Fig.l. The filter 300 includes a precombinator 302, a regulator 304, an adder 306, an accumulator 308 and a quantizer 310. The precombinator 302 is connected to the pilot channel adder 152 (Fig. 1) and receives reduced pilot symbols at a predetermined rate, for ex. 19.2 KHz. The precombinator 302 combines the pilot symbols received successively to form combined pilot symbols. This acts to reduce the memory requirements of the filter 300. For example, the precombinator can add two pilot symbols, designated p (n) and p (n + l) to produce a combined pilot symbol, which is then stored. In applications where requirements are not a concern, the precombinator can be omitted.

The precombinator 302 displaces the combined pilot symbols in sequence in the regulator. The controller preferably stores 21 combined pilot symbols, corresponding to the 42 pilot symbols received from the pilot channel adder 152. This also corresponds to a group delay of 1.1 milliseconds.

During each period of combined pilot symbols, the regulator 304 moves a new combined pilot symbol on the regulator 304 and moves the older combined pilot symbol out of the regulator 304. The adder 306 adds the content of the regulator to the new combined pilot symbol provided by the precombinder 302 to adder 306. The sum is accumulated in accumulator 308. The sum is then quantized to reduce the complexity of the circuit. The quantified result corresponds to the estimation of the channel phase and channel gain.

As noted, the filter 300 is characterized by a group delay, preferably equal to 21 pilot symbols or 1.1 milliseconds. In accordance with this invention, if the filter 300 is used to provide the filtering function of the filter 154 (Fig. 1), the delay introduced by the delay element 162 is essentially equal to the group delay of the filter 300.

As shown, the performance of the near-optimal DS-CDMA downlink receiver can be achieved by using a low pass filter to jointly estimate phase and channel gain. To achieve this near-optimal performance, it is necessary to allow a demodulation delay in the order of one to two milliseconds. While such a modest delay is tolerable for voice communication, it may be undesirable for the detection and demodulation of the power control indicator transmitted from the base station and received in the mobile station as a power control allocator. For example, the IS-95 Specification of TIA / EIA, which defines the standard DS-CDMA, requires that the output power of the mobile station be established within 0.3 dB of its final value within 500 microseconds of the reception of the power control bits by the mobile station. Accordingly, separate demodulation is necessary for the power control indicator.

In order to reduce the delay in detecting the power control indicator without sacrificing traffic channel performance, this invention separates the demodulation of the power control indicator from the demodulation of the traffic channel signals. More specifically, this invention employs two separate demodulators, one for the demodulation of the power control indicator with little or no demodulation delay and the other with longer delay appropriate for the demodulation of traffic channel signals, as described further above.

Therefore, the method according to this invention includes together estimating a complex representation of a traffic channel phase and a traffic channel gain and separately estimating a complex representation of a power control channel phase and a gain. of power control channel. The traffic channel signals will be demodulated using the traffic channel phase and the traffic channel gain. The power control allocator is demodulated with the power control channel phase and the power control channel gain.

This approach is feasible because, with reference to a DS-SDMA system, the power control bits are uncoded and the error ratio curve for an unencrypted signal is generally quite flat in the noise / signal ratio scale that we are interested. As a result, this invention uses an estimator with little or no delay for the demodulation and detection of power control bits. The error ratio of the power control allocator generated by the use of this zero or very short delay channel estimator is only slightly lower than the error rate generated by the use of a near-optimal estimator with sufficient delay. Also, the performance of the uplink receiver (i.e. the receiver in the base station receiving transmissions from the mobile station incorporating a receiver in accordance with this invention) is not very sensitive to the error rate of the control allocator. power. Therefore, the performance of the communication channel is not significantly degraded due to the use of the zero or short delay estimator.

Although the demodulated power control signal and the demodulated traffic signal have different delays, these delays are preferably fixed and known. Accordingly, there is no confusion about the nature of the demodulated signal received in the combiner 118 (Fig. 1). Also, although it is necessary to implement two separate channel estimators for this invention, the complexity of a receiver according to this invention does not increase substantially with respect to the prior art implementation. Three possible embodiments are illustrated in Figs. -6.

With reference to Fig. 4, there is shown a first alternative block diagram of a receiver circuit 400 for use in the mobile station 100 of Fig. 1. The receiver circuit 400 can be used as an indicator of a tilt receiver circuit, as illustrated in FIG. 1, to demodulate DS-CDMA signals and other spread spectrum communication signals. The receiver circuit 400 is configured to be connected to an antenna 402 and includes a filter circuit 404, a reducer 405, a pilot channel decoder 406, a first channel estimator 408, a second channel estimator 410, a first channel generator conjugate 412 and a second conjugate generator 414. The receiver circuit 400 further includes a traffic channel decoder 416, a switch 417, a short delay element 418, a delay element 420, a traffic channel demodulator 422 and a demodulator of power control 424.

In operation, the spread spectrum signals are transmitted by a remote transmitter through a communication channel and detected by the antenna 402. The spread spectrum signals are processed by the filter circuit 404, as described above with respect to the Fig.l. In the reducer 405, a reducing code is applied as a short pseudorandom (PN) noise code to the received spread spectrum signals. The PN code is used in the reducer 405 to reduce the signal in the receiver. Reducer 405 produces a reduced signal.

The reduced signal is transmitted to the pilot channel decoder 406. The pilot channel decoder 406 applies a code, for example a Walsh code, to the reduced signal to decode the signal and the decoded signal is summed to produce pilot symbols. The code is common to all users in the system so that all users can decode the pilot channel. The pilot channel can include, for example, data comprising all logic 1 to allow the • determination of the phase and gain of the communication channel. In applications such as the DS-CDMA system in accordance with IS-95, where the pilot channel is encoded using the Walsh code (0), the function of applying this Walsh code to the reduced signal in the channel decoder 406 pilot can be omitted. The pilot channel decoder 406 produces pilot symbols. The pilot symbols are provided to the first channel estimator 408 and the second channel estimator 410.

The first channel estimator 408 estimates the channel phase and channel gain for a traffic channel, yielding a first estimated channel gain and a first estimated channel phase. The first channel estimator 408 may be implemented as a low pass filter, as described above in connection with Figs. 2-3, or in any other suitable form. For example, a finite impulse response filter of the fourth order (IIR) with a delay of 1.5 milliseconds gives a near-optimal performance. A response filter of 61 finite shunt pulses (FIR) gives a similar performance with approximately the same delay.

The first channel estimator 408 produces a complex number having a magnitude and a phase and contains information corresponding to the channel phase and the channel gain. This complex number is provided to the first conjugate generator 412 which determines the complex conjugate of the complex number. The conjugate of the complex number is provided to the traffic channel demodulator 422.

The second channel estimator 410 erases the channel phase and the channel gain for the power control indicator or the power control bits, yielding a second estimated channel gain and a second estimated channel phase. The second channel estimator 410 can be implemented as a low pass filter. If the IIR estimators are used to implement the receiver circuit 400, it is more efficient to use a separate pole IIR filter as an estimator for the power control bits. In this implementation, the second IIR filter must be evaluated for each pilot symbol. This increases the computational complexity, but only slightly because this estimator is very simple. An alternative embodiment of the second estimator will be described below with Fig.6.

The second channel estimator 410 produces a complex number having a magnitude and a phase and contains information corresponding to the channel phase and the channel gain for the power control allocator. The complex number is provided to the second conjugate generator 414 which determines the complex conjugate of the complex number. The conjugate of the number comlejo is provided to the power control demodulator 424.

The reduced signal is also transmitted to the traffic channel decoder 416. The reduced signal contains both traffic data and a power control allocator. The traffic data correspond to information, for example voice or data, transmitted from a remote transmitter on the channel to the receiver circuit 400. The traffic data is encoded. The power control allocator corresponds to power control information transmitted from the remote transmitter to the receiver circuit to control the transmit power of a transmitter associated with the receiver circuit, for example the transmitter 126 (Fig.l). The power control allocator is not convolutionally encoded. The traffic channel decoder applies a traffic code, for example a Walsh code, to the reduced signal to decode the signal. The traffic code is unique to the receiver circuit 400 so that other users in a system including the receiver circuit 400 can not decode the signal. The traffic channel decoder 416 produces traffic symbols that correspond to both the control power allocator and the traffic data. The traffic symbols that correspond to the power control indicator are referred to as power control symbols.

The switch 417 selectively supplies the traffic symbols to the short delay element 418 or to the delay element 420. When the traffic symbols correspond to the power control allocator, the switch 417 provides the traffic symbols to the short delay element 418. When the traffic symbols correspond to traffic data, the switch 417 provides the traffic symbols to the delay element 420.

The delay element 420 delays the traffic symbols for a predetermined time by producing a delayed traffic channel signal comprising delayed traffic symbols. The short delay element 418 delays the power control symbols for a predetermined second time to produce a delayed power control allocator. The delay element 420 can be implemented as a first in first out (FIFO) regulator that sets the first predetermined time during which the traffic symbols will be delayed. Similarly, the short delay element 418 can be implemented as a FIFO regulator which sets the second predetermined time during which the traffic symbols will be delayed.

According to this invention, the second predetermined time is less than the first predetermined time. The first channel estimator 408 is characterized by a first group delay. Similarly, the second channel estimator is characterized by a second group delay. The second group delay is preferably shorter than the first group delay. The first predetermined time is set essentially equal to the first group delay. Similarly, the second predetermined time is set essentially equal to the second group delay. The second predetermined time is preferably less than 500 microseconds or the short delay element 418 may be omitted to reduce the complexity of the receiver design while simultaneously maintaining the proper accuracy of performance.

The short delay element 418 transmits the delayed power control dispatcher to the power control demodulator 424. The delay element 420 transmits the delayed traffic symbols to the traffic channel demodulator 422. The traffic channel demodulator 422 and the power control demodulator 424 are preferably implemented as multipliers. The power control demodulator 424 multiplies the delayed power control allocator by the complex conjugate of the complex representation of the channel phase and the received channel gain from the second conjugate generator 414. The traffic channel demodulator 422 multiplies the traffic symbols delayed by the complex conjugate of the complex representation of the channel phase and the channel gain received from the conjugate generator 412. The demodulated power control allocator and the demodulated traffic symbols are then available for a new processing, as in the combiner 118 (Fig.l).

Fig.5 shows a block diagram of operation of a second alternative receiver circuit 500. The receiver circuit 500 is configured to be connected to an antenna 502 and includes a filter circuit 504, a reducer 505, a decoder of pilot cnaal 506 , a causal filter portion 508, an anti-causal filter portion 510, an adder 512, a delay element 514, a first conjugate generator 516 and a second conjugate generator 518. The receiver circuit 500 also includes a channel decoder. traffic 520, a switch 521, a short delay element 522, a delay element 524, a power control demodulator 526 and a traffic channel demodulator 528. The operation of the receiver circuit 500 is able to detect, reduce, decode and demodulate an extended spectrum communication signal having a traffic channel, pilot channel and power control allocator, generally coincides with the operation of the receiving circuit 400 illustrated in Fig. 4, with variants described below.

The causal filter portion 508 and the anti-causal filter portion 510 together form an FIR filter. In receiver circuit 500, the pilot channel decoder 506 provides pilot symbols to the causal filter portion 508 of the FIR filter, including the center coefficient. In response, the causal filter portion 508 generates a causal output. The pilot channel decoder 506 also provides the pilot symbols to the anti-causal filter portion 510 of the FIR filter. In response, the anti-causal filter portion 510 generates an anti-causal output.

The causal output is used as an early channel estimate for the demodulation of the power control allocator without aggregate delay. The causal filter portion 508 provides the causal output to the second conjugate generator 518 for the generation of the complex conjugate of the causal output. The complex conjugate is provided to the power control demodulator 526. The power control demodulator 526 multiplies the complex conjugate and the power control symbols received from the traffic channel decoder 520, by demodulating the power control symbols.

The causal output is also provided to the delay element 514, which retards the causal output for a predetermined time, which is preferably equal to the length of the anti-causal filter, to produce a delayed causal output. The adder 512 produces a final channel estimate by adding the delayed causal output and the anti-causal output. The adder provides the final channel estimate for the conjugate generator 516 for the generation of a complex conjugate. The conjugate is provided to the traffic channel demodulator 528. The traffic channel demodulator 528 multiplies the complex conjugate and the traffic symbols received from the channel decoder 520, by demodulating the traffic symbols.

Alternatively, if a short delay can be tolerated for the demodulation of the control bits, the causal filter portion should include some, say M, anticausal coefficients. In this case, the power control symbols should be delayed by M symbols before being demodulated using the current early delay channel estimation.

Fig. 6 is a block diagram of operation of a power control channel estimator 600 for use in the receiver circuit 400 of Fig. 4. The channel estimator 600 is in the form of a low pass filter that exponentially averages successive values of the pilot symbols to produce the complex representation of the estimated channel gain and the estimated channel phase. The channel estimator 600 includes a shifter 602, an adder 604, a shifter 606, an adder 608, a delay element 610 and a quantizer 612. The channel estimator 600 receives the pilot symbols from the pilot channel decoder 406. The Pilot symbol is in the form of an 8-bit binary value. The displacer 602 moves a pilot symbol 2 bits to the left, forming a value of 10 bits. The adder 604 adds the current pilot symbol to the delayed pilot symbol received from the delay element 610. The sum, a binary value of 11 bits, is shifted 3 bits to the right by the displacer 606. The adder 608 adds this value to the symbol delayed pilot received from the delay element 610, producing a value of 10 bits. The sum is provided to the delay element 610 to process it with the subsequent pilot symbol. The sum is also provided to the quantizer 612, which keeps the 8 most significant bits as the channel estimate.

As can be seen from the foregoing, this invention provides a method and apparatus for demodulating an extended spectrum communication signal, including power control bits. The channel phase and the channel gain are estimated together by averaging or filtering the pilot symbols with a low pass filter. The traffic symbols are slightly delayed to adapt to the filtering delay. The power control bits are not delayed or delayed for only a short time to ensure detection and response to the power control bits within a specified period of time. The inventors have determined that the joint estimation method according to this invention provides near optimal estimates for both the phase and the channel gain. The DS-CDMA receiver implemented in accordance with this invention provided a frame error ratio (FER) 0, 7 to 0.9 dB better than the conventional design that smocks a locked phase loop for the channel phase estimation and a separate channel gain estimator. The FER is only 0.15 dB displaced from the result obtained using the perfect channel phase and channel information to demodulate traffic symbols. In addition, this invention can be easily implemented in hardware or software or a combination of both. Also, this invention allows the DS-CDMA power control bits to be detected within the specified time period of 500 microseconds. While a particular embodiment of this invention has been shown and described, modifications may be made to it. For example, the filter used to estimate the channel phase and channel gain can be implemented using FIR and IIR techniques. The level of accuracy of the estimate can be adapted to the accepted level of complexity of the filter. The filters that produce the first and second complex channel estimates can be essentially the same. Also, it will be recognized that the operating elements of the receiver circuit 400, receiver circuit 500 and channel estimator can be implemented in hardware, in software or in any combination of both that increases the efficiency and performance of the design. Therefore it is our intention that the appended claims cover all those changes and modifications that fall within the true spirit and scope of the invention.

Claims (9)

RETVINDICATIONS

1. A method for demodulating an extended spectrum communication signal, the method characterized by: receiving (150) the spread spectrum communication signal over a communication channel; detecting (151) a pilot channel signal in the spread spectrum communication signal, producing pilot symbols (166); producing (408 an estimated channel gain and an estimated channel phase for the communications channel; detecting (158) in the extended spectrum communication signal a traffic channel signal, producing traffic symbols (170); ) the traffic symbols for a predetermined time, producing delayed traffic symbols, and demodulating (164) the delayed traffic symbols using the estimated channel gain and the estimated channel phase.

2. The method of claim 1 further characterized by: producing (416) a power control allocator in response to the pilot channel signal; delaying (418) the power control allocator a predetermined time to produce a delayed power control dispatcher, the second predetermined time is different from the first predetermined time; and demodulating (424) the allocator of power control using the estimated channel gain and the estimated channel phase.

3. The method of claim 2 characterized also in that the step of producing an estimated channel gain and an estimated channel phase comprises the steps of filtering (408) the pilot symbols to obtain a first estimated channel gain and a first estimated channel phase , and filter (410) the pilot symbols to obtain a second estimated channel gain and a second estimated channel phase and because the step of demodulating the delayed traffic symbols comprises using the first estimated channel gain and the first estimated channel phase and because the step of demodulating the delayed power control allocator comprises using the second estimated channel gain and the second estimated channel phase.

4. The method of claim 2 also characterized in that the second predetermined time is essentially zero milliseconds.

5. The method of claim 4 also characterized in that the first predetermined time is on a scale of 0.5 to 2 milliseconds.

6. The method of claim 1 further characterized in that the first predetermined time is in the range of 0.5 to 2 milliseconds.

7. The method of claim 1 further characterized by the step of delaying (162) only the traffic channel signal to produce the delayed traffic symbols.

8. The method of claim 1 further characterized in that the step of producing an estimated channel gain and an estimated channel phase comprises the step of filtering the pilot symbols by a low pass filter (154) to obtain the estimated channel gain and the phase of estimated channel.

9. A method according to claim 8 characterized also in that the step of producing an estimated channel gain and an estimated channel phase also comprises the step, subsequent to the step of filtering by low pass filter, generating a complex conjugate (156) to produce the estimated channel phase and the estimated channel gain.