WO1999052249A1

WO1999052249A1 - A receiver for spread spectrum communications signals

Info

Publication number: WO1999052249A1
Application number: PCT/EP1999/001973
Authority: WO
Inventors: Nicholas William Anderson; Rorie O'neill; Reza Karimi
Original assignee: Motorola Limited
Priority date: 1998-04-07
Filing date: 1999-03-22
Publication date: 1999-10-14
Also published as: GB9807335D0

Abstract

A signal processor (7,8) for a telecommunications receiver, and particularly applicable to TD-CDMA systems, provides a means for removing interference from a received data block (2) caused by the multi-path effects of the midamble portion (3) of the received signal. The known midamble sequence (3) is convolved with an estimate of the propagation channel (5) and the result is subtracted (8) from the received signal (4). The corrected signal can then be input to a conventional joint detection processor (6) for further processing.

Description

A RECEIVER FOR SPREAD SPECTRUM COMMUNICATIONS SIGNALS

This invention relates to radio receivers and particularly to radio receivers for TDMA or TD-CDMA Mobile Communications Systems.

In a TDMA (Time Division Multiple Access) system, a communication channel consists of a time slot in a periodic train of time intervals over the same frequency. Each period of time slots is called a frame. A given signal's energy is confined to one of these time slots. Adjacent channel interference is limited by the use of a time gate or other synchronisation element in the receiver that only passes signal energy received at the proper time.

In contrast to TDMA, a CDMA (Code Division Multiple Access) system allows signals to overlap in both time and frequency. In principle, in a CDMA system the information data stream to be transmitted is impressed upon a much higher rate data stream known as a signature sequence. Typically, the signature sequence data is binary, providing a bit stream. One way to generate this signature sequence is with a pseudo-noise (PN) process that appears random, but can be replicated by an authorised receiver. The information data stream and the high bit rate signature sequence stream are combined by multiplying the two bit streams together, assuming the binary values of the two bit streams are represented by +1 or -1. This combination of the higher bit rate signal with the lower bit rate data stream is called coding or spreading the information data stream signal. Each information data stream or user channel is allocated a unique spreading code.

One CDMA technique "traditional CDMA with direct spreading", uses a signature sequence to represent one bit of information. Receiving the transmitted sequence or its complement (the transmitted binary sequence values) indicates whether the information bit is a 0 or a 1. The signature sequence usually comprises n-bits and each bit is called a "chip". The entire n-chip sequence, or its complement is referred to as a transmitted 2

symbol. The receiver correlates the received signal with the known signature sequence of its own signature sequence generator to produce a normalised value ranging from -1 to +1. When a large positive correlation results, a 0 is detected. When a large negative correlation results, a 1 is detected.

A plurality of coded information signals modulate a radio frequency carrier, for example, quadrature phase shift keying (QPSK) and are jointly received as a composite signal at a receiver. Each of the coded signals overlaps all of the other coded signals, as well as noise-related signals in both frequency and time. If the receiver is authorised, then the composite signal is correlated with one of the unique codes, and the corresponding information signal can be isolated and decoded.

In a TD-CDMA (Time Division-Code Division Multiple Access) system, each time-slot within the TDMA time frame is divided into a plurality of channels which are uniquely distinguished from each other by means of the CDMA process. The use of such a system gives some advantages over a communications system using either TDMA or CDMA alone.

A TD-CDMA receiver needs to be able to separate the time slots in each frame and also separate the coded signals (by using a correlation process). By such means, channels can be separated at the receiver. If all the unique codes within a time slot are orthogonal, rejection of unwanted channels at the receiver is complete. However, the transmitted composite signal can become corrupted in the transmission process and in such cases, the codes arriving at the receiver are no longer orthogonal. This results in signals from unwanted channels corrupting the data on the wanted channel.

This form of interference can be removed by the known process of "joint detection". In a typical joint detection receiver, the receiver is provided with all the codes and therefore able to decode all the channels within a given time slot. Further processing of the channel outputs enables a better estimate of the wanted channel signal to be obtained. One known method of estimating an unknown transmitted channel symbol sequence in a joint detection receiver involves zero-forcing block-linear equalisation. The current TD-CDMA proposal for the air interface of the third generation mobile telecommunications systems (UMTS-Universal Mobile Telecommunications Systems) requires that joint detection of multiple codes is performed by the receiving end of the link. This requires that each mobile radio channel from the one (in the case of the downlink) or multiple (in the case of the uplink) transmitters to the receiver must be estimated.

The propagation channel through which a transmitted burst passes, comprises a transmit filter, a mobile radio channel and a receiver filter. The propagation channel is of a fixed length - W chips. A burst typically comprises at least one data block of say, N symbols and a pre-defined training sequence.

Normally, the joint detection equations can be solved using propagation channel estimates and knowledge of the spreading codes used.

The signal model normally assumed is r=Ad where r is the received signal vector (of length NQ+W-1 chips) arising due to a symbol block of length N Symbols, where Q is the spreading factor. d is the vector containing the symbol transmitted for each code and

A is the modulation matrix containing information regarding the user spreading codes and channel estimates. This is of size (NQ+W-1) NK where K is the number of codes (or user channels).

These equations however as they stand, do not take into account the interference caused by the training sequence portion of the burst due to reflections of the transmitted signal prior to their arrival at the receiver, i.e. multi-path effects are not taken into account. The equations are not representative of the true situation for the last W-l chips of r when processing the first block, nor for the first W-l chips of r when processing the second block. This results in a performance degradation because the joint detection processing does not take the interference caused by the training sequence into account. 4

This invention aims to remove such interference.

The training sequences, often referred to as midamble sequences, in a typical TD-CDMA mobile communications system are known. Also known are the relevant channel coefficients. By using these two facts, the interference caused by a dispersive propagation channel acting on the midamble can be removed from the datablock portions of the received signal.

Accordingly, the present invention consists of a signal processor for a receiver forming part of a telecommunications system in which an information signal burst transmitted from a remote transmitter and comprising at least one data portion and a training sequence portion is received at an input of the receiver via a propagation channel, said burst being degraded in the transmission process, wherein the signal processor includes apparatus for removing degradation from the received burst incurred in the transmission process, said apparatus comprising a convolver for convolving an estimate of the propagation channel characteristics with the training sequence of the received burst to generate a correction signal, and a subtracter for subtracting the correction signal from the received burst to produce a modified burst signal for further processing.

The further processing can comprise a joint detection processing arrangement.

The signal processor may be incorporated in the receiving circuitry of a base station and/or mobile station of a mobile communications system.

An embodiment of the invention will now be described, by way of example only, with reference to the drawings of which;

Figure la and lb are schematic diagrams showing the structure of a transmitted and received burst, respectively, in a telecommunications system to which the invention is particularly applicable, Figure 2 is a schematic block diagram of a receiver incorporating apparatus in accordance with the invention.

In Figure la, a transmitted TD-CDMA burst comprises a first datablock 1 comprising N data symbols and a second data block 2 comprises N symbols, the first and second datab locks 1,2 being separated by a midamble portion 3 of L chips. By virtue of the CDMA technique previously explained, a single burst comprises information associated with a plurality of communications channels.

A propagation channel through which each burst passes typically comprises a transmit filter (not shown), the air interface and the receiver filter (not shown) and is defined as a having a length of W chips.

Figure lb shows the received burst in which part of block 1 (length W-l chips) has shifted into the midamble time slot and part of a midamble 3 (of length W-l chips) has shifted into the second datablock 2 timeslot. This shifting is due to reflections of the transmitted burst over the air interface being detecting at the receiver (i.e. multi-path effects).

The resulting inter-symbol interference caused by the midamble portion being received during the second block timeslot is removed by the apparatus incorporated in Figure 2.

Referring then to Figure 2, a telecommunications receiver, installed in a base station (not shown) and a mobile station (not shown) of a mobile telecommunications network, for receiving a TD-CDMA burst on line 4 is provided with a channel estimator 5 whose output is connected to a joint detection processor 6 and to a first input of a convolver 7. A second input of the convolver 7 is provided with the known midamble sequence. An output of the convolver 7 is fed into a subtracter 8 which also receives the burst on line 4. The output of the subtracter 8 is fed to a joint detection processor 6.

From the received burst an estimate of the channel characteristics is made in the known manner by the channel estimator 5. This estimate is then convolved in the convolver 7 6

with the known midamble sequence and the result is subtracted (by the subtracter 8) from the received burst. The resulting signal is input to the joint detection processor 6 for processing in accordance with known techniques. The output of the joint detection 6 comprises estimates of the transmitted symbols for each user channel.

For block 1, the last W-l chips contain a contribution from the midamble. This contribution is simply the convolution of the first W-l chips of the midamble with the propagation channel. This convolution is performed in the convolver 7 and the interfering signal is subtracted from the last W -l chips ofr. Similarly, for block 2, the interference caused by the midamble 3 is simply the convoltion of the last W-l chips of the midamble sequence with the propagation channel for block 2, this can be subtracted from the first W -1 chips ofr in order to eliminate the interference caused by the midamble 3.

In this manner, 2x(W-l)/Q (rounded up) symbols in each burst will have increased likelihood of being estimated correctly, thus improving performance through reduced bit error rate, or allowing for reduced transmit power to maintain the same bit error rate as was achieved without this additional processing.

Since the number of symbols that are adversely affected by the midamble increases with the mobile radio channel link (or the transmit/receive filter impulse response duration), so the improvement gain through the use of this method also increases. For long mobile radio channels, e.g. hilly terrain, considerable improvements in performance can be achieved.

Also, on the uplink, mobile users will be received with different relative delays so even if actual channel dispersion is low, it can appear as if a given user has a significant impulse response duration, i.e. the delay must be absorbed by the channel estimation process. Thus the invention improves performance in this situation also.

While this invention is most useful when applied to the TD-CDMA case, it is also applicable to TDMA systems.

Claims

7CLAIMS

1. A signal processor (7/8) for a receiver forming part of a telecommunications system in which an information signal burst transmitted from a remote transmitter and comprising at least one data portion (2) and a training sequence portion (3) is received at an input of the receiver via a propagation channel, said burst being degraded in the transmission process, wherein the signal processor includes apparatus for removing degradation from the received burst incurred in the transmission process, said apparatus comprising a convolver (7) for convolving an estimate of the propagation channel's characteristics with the training sequence (3) of the received burst to generate a correction signal, and a subtracter (8) for subtracting the correction signal from the received burst to produce a modified burst signal for further processing.

2. In a telecommunications system in which an information signal burst(l, 2, 3) is transmitted from a remote transmitter to a receiver via a propagation channel, said burst comprising at least one data portion (2) and a training sequence portion (3), and wherein said burst suffers degradation in the transmission process, a method for removing said degradation including the steps of: receiving the transmitted burst, in the receiver, convolving an estimate of the propagation channel's characteristics with the training sequence portion to generate a correction signal, and subtracting the correction signal from the received burst to produce a modified undegraded burst .