Lucent Technologies Inc.

Agere Systems Inc.

Alcatel Lucent

Markus Rupp

Harish Viswanathan

Shankar Narayanaswamy

Rajeev Krishnamoorthy

Xiaodong Li

Jaiganesh Balakrishnan

Hui-Ling Lou

Eric C Beck

Use of a training sequence having terms that are orthogonal to each other are employed to considerably speed up execution of the LMS algorithm. Such orthogonal sequences are developed for a channel that is described as a finite impulse response (FIR) filter having a length Mfrom the already existing orthogonal training sequences for at least two channels that have respective lengths Mand Meach that is less than Msuch that the product of Mand Mis equal to Mwhen Mand Mhave no common prime number factor. More specifically, a set of initial existing orthogonal training sequences is found, e.g., using those that were known in the prior art or by performing a computer search over known symbol constellations given a channel of length M. Thereafter, an orthogonal training sequence of length Mis developed, where the product of Mand Mis equal to Mby repeating the training sequence old1 Mnumber of times to form a first concatenated sequence and repeating the training sequence old2 Mnumber of times to form a second concatenated sequence, so that both the first concatenated sequence and the second concatenated sequence have the same length. Each term of the first concatenated sequence is multiplied by the correspondingly located term in the second concatenated sequence which is placed in the same location in a new sequence made up of the resulting Mproducts. This new sequence is an orthogonal sequence of length M.

Training sequence for low latency LMS implementation

A channel sounding system employs orthogonal sequences to meet the Cramer-Rao bound in estimating the channel and achieves considerable simplification of the structure necessary to perform the channel sounding. These advantages are achieved by developing orthogonal sequences of substantially arbitrary length as a function of first and second existing orthogonal sequences and using such orthogonal sequences for channel sounding in lieu of M-sequences. The techniques of the invention are especially suited to systems that use multiple antennas at the transmitter and multiple antennas at the receiver, so called multiple-input multiple-output (MIMO) systems.

Optimal channel sounding system

Orthogonal sequences can be developed and used for use training and synchronizing in CDMA and TDMA systems. In particular, once a sequence is developed that has the length of the product of the channel length and the number of transmit antennas, the sequence is offset by a different amount for each transmit antenna. For example, each sequence could be offset by a multiple of the channel length for each transmit antenna, where the multiple ranges from 0 to N&#x2212;1, where N is the number of transmit antennas. Furthermore, by not using exactly the same amount of offset shifting for each transmit antenna, e.g., not having each signal offset by the channel length, but keeping the overall total shifting the same, e.g., the average of each shift is the channel length, it is possible to determine at a receiver from which transmit antenna a particular signal originated.

Training and synchronization sequences for wireless systems with multiple transmit and receive antennas used in CDMA or TDMA systems

Signal processing operations are performed in a digital communication system receiver on a sequence of received symbols, each representing a number of information bits. The symbols correspond to points in a given modulation constellation generated by applying a predetermined rotation, e.g., a 45&#xb0; rotation, to an otherwise conventional modulation constellation, e.g., a QPSK constellation, a 16-QAM constellation, etc. The use of the rotated constellation allows certain signal processing operations, such as filtering, Least-Mean-Squares (LMS) estimation, and Maximum-Likelihood (ML) sequence detection via the Viterbi algorithm, to be performed without the need for multipliers. By eliminating or substantially reducing the number of required multiplication operations, the invention significantly reduces the complexity and delay associated with the corresponding signal processing circuitry. Advantageously, this reduction in complexity and delay is accomplished without the use of any approximation or other reduction in precision.

Multiplier-free methods and apparatus for signal processing in a digital communication system

In a TDMA system a multifunction control channel is employed that carries data used for implementing channel access functionality for users and is also employed at the remote terminals in the performance of modem functionality. In other words, information necessary to accurately receive data on the user traffic channels is obtainable by processing the multifunction control channel. This is achieved by arranging the broadband channel of the TDMA system as a repeating frame having time slots in which at least one time slot is reserved for use as the multifunction control channel, and the data that is transmitted on the multifunction control channel is encoded using a spread spectrum format. To this end the data carried on the multifunction control channel is encoded in a way that uses each symbol of the multifunction control channel as if it were a so-called spread spectrum 'chip'. At least one of the remaining time slots of the frame, and preferably all of the remaining time slots, which are typically used to carry user traffic arranged as logical channels, are not spread spectrum encoded. With regard to modem functionality, the remote terminals use the received multifunction control channel to a) perform frame synchronization, i.e., frame timing, b) perform frequency offset estimation, c) obtain an estimate of the channel impulse response, and d) estimate the received signal strength for use in gain control.

Control channel for time division multiple access systems

In a wireless communication system, the constellation mapping scheme employed may be changed on a per-time-slot basis, i.e., from time slot to time slot, so that the constellation used to encode the symbols of each time slot may be different for each time slot within a single frame and may be different for a particular time slot in different consecutive frames. In other words, several constellation mapping schemes are available, with each providing the ability to transmit a different number of bits per symbol, and the particular constellation mapping scheme employed for any time slot need be selected for that time slot only. The ability to use any particular constellation mapping scheme is dependent on the current channel quality. The particular constellation mapping used for the user data of the time slot may be indicated in the preamble of the time slot. To this end, each time slot may have its own preamble that is mapped with a constellation mapping scheme which is a) known &agrave; priori, b) may be the same for all time slots, and c) may be different from the constellation mapping scheme used to encode user data in the time slot. A receiver can determine the constellation mapping used for each time slot from solely from the preamble of the time slot.

Demodulation method for receiver

In a system in which the modulation scheme employed may change on a per-time-slot basis, a user's available data rate may change even if the user maintains the same number of time slots, because the capacity of the user's time slots may change. If one or more such changes in modulation scheme occurs, the data rate available to the user may no longer match his current data rate needs. Therefore, a method is disclosed for determining the number of time slots to assign to a user as a function of the user's current data rate requirement and the actual current cumulative data rate of the time slots currently assigned to the user. The user data rate requirement may further be a function of a quality of service (QoS) contracted for by the user. Thus, the assignment of the time slots within the frame is made dynamically and the number of time slots employed by a user is increased or decreased according to the user's current data rate requirements and the ability of the time slots already assigned to the user to meet those requirements.

Medium allocation method

A two layered segmentation technique coupled with a two layer error detection technique is used to implement the wired to wireless network interface. A wireless network divides network layer packets from a wired network into radio data link packets, and the information within the radio data link packets is divided into portions that can be placed into time slots. Error detection is performed on each of the time slots and also at the radio data link packet level to determine if there is an error within the radio data link packet. Errors detected in any radio data link packet only requires retransmission of the radio data link packet in which the error was detected, and, advantageously, do not require retransmission of the entire network layer packet as would have been required in a system that mapped directly from network layer packets to time slots. Further advantageously, the system is able to be employed by systems that utilize dynamic constellation mapping schemes which result in different time slots for the same user being mapped with different constellations, and so they have different bit to symbol ratios. This is because such changes in the constellation mapping scheme are handled at the time slot level, and are not seen at the radio data link packet level. The segmentation of the network layer packets into radio link packets is independent of the number and size of the time slots which will carry the radio link packets.

Data link protocol for wireless systems

In the method of pulse shaping according to the present invention, a set of bits representing a symbol is received, and an output value is generated by adding or subtracting a received coefficient and a received value based on a first predetermined one of the set of bits. The generated output value is then supplied as either a real or imaginary output value based on at least a second predetermined one of the set of bits.

Method and apparatus for pulse shaping

In a wireless communication system, the constellation mapping scheme employed may be changed on a per-time-slot basis, i.e., from time slot to time slot, so that the constellation used to encode the symbols of each time slot may be different for each time slot within a single frame and may be different for a particular time slot in different consecutive frames. In other words, several constellation mapping schemes are available, with each providing the ability to transmit a different number of bits per symbol, and the particular constellation mapping scheme employed for any time slot need be selected for that time slot only. The ability to use any particular constellation mapping scheme is dependent on the current channel quality. The particular constellation mapping used for the user data of the time slot may be indicated in the preamble of the time slot. To this end, each time slot may have its own preamble that is mapped with a constellation mapping scheme which is a) known a priori, b) may be the same for all time slots, and c) may be different from the constellation mapping scheme used to encode user data in the time slot. A receiver can determine the constellation mapping used for each time slot from solely from the preamble of the time slot.

Modulation method for transmitter

A method and apparatus for equalizing a signal to accurately deconvolute the channel includes deriving a functional relationship between Doppler frequency and variance between the instantaneous energy of the received signal and the mean energy of the received signal. Theoretical values of Doppler frequency for a range of values of variance are mapped based on the derived functional relationship. The variance of a received signal is determined and the corresponding Doppler frequency estimate selected from the map. Optimal equalization learning parameters are then selected based on the corresponding Doppler frequency estimate so that the received signal may be deconvoluted independent of the impact of Doppler frequency.

Apparatus and method for equalizing a signal independent of the impact of doppler frequency

An effective receiver is achieved for situations where the receiver is mobile, and may be traveling at relatively high speed, and where the receiver's internal demodulation oscillator causes a frequency offset, with processing that shares a common algorithm for both frequency offset and channel characteristics estimations. Specifically, the commonly employed algorithm, such as the LMS algorithm, computes an estimate of the frequency offset, and that very same algorithm is also used to estimate the channel characteristics. When the LMS algorithm is used, a frequency offset estimate can be derived from signals derived in the course of executing the LMS algorithm. A frequency compensation factor is then developed and applied to the incoming signal to create a signal that that not have an appreciable frequency offset. That signal is then applied to a process that also employs the LMS algorithm, in combination with a detection algorithm, such as, for example, the Viterbi algorithm, to recover from the incoming signal the information signals that had been encoded into the incoming signals.

Channel tracking in a mobile receiver

A wireless communication terminal comprises a first housing section including an antenna for enabling wireless communications with the terminal and a second housing section including a second communications antenna, the second housing section being extendibly connected to the first housing in an manner such that when the second housing section is in its extended position the second antenna is located one half wavelength from the first antenna. One or both of the antennas may be a fixed type, e.g., a whip or stub, or a retractable type, e.g., telescoping. The antennas may be located external or may be internal (e.g., patch type antennas) to the housing sections.

Growing community of inventors

Lincroft, NJ, United States of America

Markus Rupp