Abstract: A timing recovery process for digital receivers employing decision-feedback equalization (DFE) or delayed decision-feedback sequence estimation (DDFSE) can be executed prior to equalization, minimizes training complexity, and provides fast equalizer start-up for transmissions of short packets. The optimum symbol timing and burst timing is determined by (i) calculating a signal-to-intersymbol interference-plus-noise ratio (SINR) index by estimating the channel impulse response and the noise power, and (ii) varying the symbol timing and the burst timing to maximize the calculated SINR index. The timing recovery process particularly improves performance of a digital receiver utilizing (a) decision-feedback equalizer structures in which the delay spread is large compared to the span of the feedforward filter, and (b) equalizer structures based on delayed decision-feedback sequence estimation in which the delay spread is large compared to the memory of the Viterbi sequence estimator section.

Abstract: Average local traffic load in a simulcast radio system is estimated to achieve efficient resource sharing and dynamic reconfiguration of the simulcast areas. A method for estimating the average local traffic load uses relative power measurements at each radio port (or distributed antenna element). The estimating method is applicable to both code-division multiple access (CDMA) and time-division multiple access (TDMA) systems. The method provides accurate estimates of the actual radio port traffic over various system conditions, while requiring only an approximate knowledge of the radio propagation environment.

Abstract: The complexity and training time of decision-feedback equalization (DFE) structures for use in broadband wireless communications are reduced. A modified DFE structure is able to pre-cancel postcursors in the received signal without requiring training of the feedback filter. In the modified DFE structure, an estimated channel impulse response is used in the feedback section. The output of the feedback section is passed to the feedforward filter to cancel the postcursors before convolving the result of the cancellation with the feedforward filter tap gains. Therefore, only the feedforward filter needs to be trained. In broadband wireless channels, the modified DFE structure can potentially speed up the training process by ten to twenty times compared to conventional structures.

Abstract: In a TDM/TDMA portable radio communications system the uplink power transmitted by the transmitter of a portable unit to a fixed port is dynamically controlled by monitoring three measures in each uplink burst received by the port. These measures are a received signal strength indicator (RSSI), a quality measure (QM), and a word error indicator (WEI). The WEI is used to adjust an RSSI threshold upward and downward. If the QM is below a predetermined QM threshold or the RSSI is below the current value of the adjustable RSSI threshold, then additional uplink power is required, unless the RSSI is greater than a predetermined maximum RSSI value. If additional power is required a power control bit (PC), included within each downlink burst transmitted to the portable, is set to ONE and the portable, in response thereto, increases its output transmitter power.

Abstract: A receiver for use in the port or portable units in a TDM/TDMA digital radio communications system is disclosed which incorporates an equalizer (107) to compensate for channel distortion due to multipath delay spread. Each received burst of information is oversampled at a multiple of the symbol rate and stored in a buffer memory (105). A joint estimator (106) processes the stored burst by operating on a known training sequence of bits within the burst to determine burst timing (where the burst actually begins) and the symbol timing (the optimum sample sample per oversampled symbol to be used for detection purposes). The estimator also determines whether the stored burst should be processed in a time-forward or time-reversed order.

Abstract: A receiver for use in the port or portable units in a TDM/TDMA digital radio communications system is disclosed which incorporates an equalizer (107) to compensate for channel distortion due to multipath delay spread. Each received burst of information is oversampled at a multiple of the symbol rate and stored in a buffer memory (105). A joint estimator (106) processes the stored burst by operating on a known training sequence of bits within the burst to determine burst timing (where the burst actually begins) and the symbol timing (the optimum sample per oversampled symbol to be used for detection purposes). The estimator also determines whether the stored burst should be processed in a time-forward or time-reversed order.

Abstract: A technique for bit synchronization and error detection of received digital data bursts in a TDM/TDMA system, such as that which will be used with low power portable digital telephony. A cyclically redundant codeword, e.g. a (161,147) codeword, is formed for transmission, using e.g. either a TDM packet or TDMA burst. The first and last bits in the codeword are then inverted to form a first set of marker bits. At a receiver, a second set of marker bits is inserted into a received word, again through inverting the first and last bits. The resulting marked word is then rotated by a pre-determined number of bits to place potentially erroneous bits at the end of this word. A multi-bit timing syndrome value is then determined and is used to access a look-up table for a value of bit slippage. The received word is advanced or retarded as specified by the bit slippage value to yield an intermediate word.

Abstract: The present invention provides two branches of a base station transmitter and a base station receiver, respectively, and inserts a delay circuit for ensuring a delay time difference between D and U waves required to obtain the best bit error rate for a DSK system, BPSK-RZ system, QPSK-RZ system or the like, into each system of the transmitter and the receiver, respectively, or into either the transmitter or the receiver of each system, thereby efficiently improving the bit error rate.

Abstract: A digital communication system, in which digital information is represented by a quaternary information symbol (11, 01, 00, 10) and the double step phase shift, each step represented by (.pi./4, 3.pi./4, 5.pi./4, 7.pi./4), is performed per 1/2 time slot in correspondence to each information symbol, so that the number of phase shift per unit time in the case of the identical transmission rate can be decreased to 1/2, as compared with a .DELTA..theta./2-DSK system, and expansion of the bandwidth can be decreased to 1/2.

Abstract: The present invention provides a digital communication system in which one time slot is divided into a first half section and a second half section and the carrier phase difference .theta. between each neighboring first half section is set to either (.pi., 0) with respect to a binary information symbol (1,0) or (0, .pi./2, .pi., 3/2.pi.) with respect to a quaternary information symbol (11, 01, 00, 10). In addition, the phase difference .psi. between the first half and the second half of any time slot is a constant value .psi..sub.0 (for example, .pi.rad) irrespective of the symbol, or .psi..sub.i determined with respect to a symbol (for example, +.pi./2 with respect to a symbol 1 or -.pi./2 with respect to a symbol 0). Therefore, an arbitrary complementary characteristic can be obtained when differentially demodulated in a multipath fading channel, irrespective of the amplitude characteristic of a demodulation signal.

Abstract: The present invention provides a communication system which employs a signal whose carrier phase is shifted by a constant amount .DELTA..theta. per time slot in correspondence to each symbol (1 or 0) transmitted, and the phase shift is performed by .DELTA..theta./2 (or -.DELTA..theta./2) in the first 1/2 time slot and further by .DELTA..theta./2 (or -.DELTA..theta./2) in the second 1/2 time slot thereby to improve the bit error rate in multipath propagation.