DE19849557A1 - Rake receiver finger tracking method especially for UMTS - Google Patents

Abstract

The method involves tracking the main finger of the rake receiver according to the change of a transmission channel using an early finger provided by an early-tracking method and a late finger provided by a late-tracking method. A reception signal is advanced with respect to the main finger at the early finger and delayed at the late finger. The energies collected from the early finger and the late finger are compared. The position of the main finger remains in dependence of the comparison or the direction of the early finger or the late finger is shifted. The direction in which the position of the main finger is shifted is determined in dependence whether the equation x=(a-c)(a+c)+(b-d)(b+d) with a exp 2 + b exp 2 indicating the energy collected by the early finger and c exp 2 + d exp 2 indicating the energy collected by the late finger is smaller or bigger than a predetermined threshold.

Description

Telecommunication systems with wireless telecommunication between mobile and / or stationary transceivers are special message systems with a message transmission link between a message source and a message sink, in which, for example, base stations and mobile parts for message processing and transmission are used as transceivers and in which

• 1. the message processing and message transmission in a preferred direction of transmission (simplex mode) or in both transmission directions (duplex operation) can be done
• 2. the message processing is preferably digital,
• 3. the transmission of messages via the long-distance transmission wirelessly on the basis of various message transmission methods FDMA (Frequency Division Multiple Access), TDMA (Time Division Multiple Access) and / or CDMA (Code Division Multiple Access) - z. B. according to radio standards such
  DECT [Digital Enhanced (formerly: European) Cordless Tele communication; see. Telecommunications Electronics 42 (1992) Jan./Feb. No. 1, Berlin, DE; U. Pilger "Structure of the DECT standard", pages 23 to 29 in connection with the ETSI publication ETS 300175-1. . . 9, October 1992 and the DECT publication of the DECT Forum, February 1997, pages 1 to 16],
  GSM [Groupe Spéciale Mobile or Global System for Mobile Communication; see. Informatik Spektrum 14 (1991) June, No. 3, Berlin, DE; A. Mann: "The GSM standard - Basis for digital European mobile radio networks", pages 137 to 152 in connection with the publication telekom praxis 4/1993, P. Smolka "GSM radio interface - elements and functions", pages 17 to 24] ,
  UMTS [Universal Mobile Telecommunication System]; see. (1): Kommunikationstechnik Electronics, Berlin 45, 1995, Issue 1, Pages 10 to 14 and Issue 2, Pages 24 to 27; P. Jung, B. Steiner: "Concept of a CDMA mobile radio system with joint detection for the third mobile radio generation"; (2): Kommunikationstechnik Electronics, Berlin 41, 1991, Issue 6, pages 223 to 227 and page 234; PW Baier, P. Jung, A. Klein: "CDMA - an inexpensive multiple access method for frequency-selective and time-variant mobile radio channels"; (3): IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, Vol. E79-A, No. 12; December 1996, pages 1930 to 1937; PW Baier, P. Jung: "CDMA Myths and Realities Revisited"; (4): IEEE Personal Communications, February 1995, pages 38 to 47; A. Urie, M. Streeton, C. Mourot: "An Advanced TDMA Mobile Access System for VMTS"; (5): telekom praxis, 5/1995, pages 9 to 14; PW Baier: "Spread spectrum technology and CDMA - an originally military technology conquered the civilian area"; (6): IEEE Personal Communications, February I 995, pages 48 to 53; PG Andermo, LM Ewerbring: "An CDMA-Based Radio Access Design for UMTS"; (7): ITG Fachberich te 124 (1993), Berlin, Offenbach: VDE Verlag ISBN 3-8007-1965-7, pages 67 to 75; Dr. T. Zimmermann, Siemens AG: "Application of CDMA in Mobile Communication"; (8): telcom report 16, (1993), volume 1, pages 38 to 41; Dr. T. Ketseoglou, Siemens AG and Dr. T. Zimmermann, Siemens AG: "Efficient subscriber access for the 3rd generation of mobile communication - multiple access procedure CDMA makes the air interface more flexible"; (9): Funkschau 6/98: R. Sietmann "Wrestling for the UMTS interface", pages 76 to 81] WACS or PACS, IS-54, IS-95, PHS, PDC etc. (see IEEE Communications Magazine, January 1995, pages 50-57; DD Falconer et al. "Time Division Multiple Access Methods for Wireless Personal Communications"].

"Message" is a superordinate term that stands for both the meaning (information) and the physical representation (signal). Despite the same meaning of a message - i.e. the same information - different signal forms can occur. So z. B. a message relating to a subject

• 1. in the form of a picture,
• 2. as a spoken word,
• 3. as a written word,
• 4. transmitted as an encrypted word or image.

The transmission type according to (1). . . (3) is usually characterized by continuous (analog) signals, wuh rend usually discontinuous in the type of transmission according to (4) Orier signals (e.g. impulses, digital signals) arise.

In the UMTS scenario (3rd mobile generation or IMT-2000) there it z. B. according to Funkschau 6/98: R. Sietmann "Wrestling for the UMTS interface", pages 76 to 81 two Partial scenarios. In a first partial scenario, the license coordinated mobile radio based on WCDMA technology (Wideband Code Division Multiple Access) based and how with GSM, operated in FDD mode (Frequency Division Duplex), while in a second sub-scenario the unlicensed unko ordinated mobile radio on a TD-CDMA technology (Time Di vision code division multiple access) and, as with DECT, operated in TDD mode (Frequency Division Duplex) becomes.

For the WCDMA / FDD operation of the universal mobile telecommunication system, the air interface of the telecommunication system contains the upward and downward direction of the telecommunication in accordance with the publication ETSI STC SMG2 UMTS-L1, Tdoc SMG2 VMTS- L1 163/98: "UTRA Physical Layer Description FDD Parts "Vers. 0.3, 1998-05-29 each have several physical channels, of which a first physical channel, the so-called Dedicated Physical Control CHannel DPCCH, and a second physical channel, the so-called Dedicated Physical Data CHannel DPDCH , with respect to their time structure (frame structure) are shown in Figs. 1 and 2.

In the downlink (radio connection from the base station to the mobile station) of the WCDMA / FDD system from ETSI or ARIB becomes the Dedicated Physical Control Channel (DPCCH) and the Dedicated Physical Data Channel (DPDCH) multiplexed in time while an I / Q multiplex takes place in the uplink, in which the DPDCH in I-channel and the DPCCH are transmitted in the Q-channel.

The DPCCH contains N _pilot pilot bits for channel estimation, N _TPC bits for fast power control and N _TFI format bits that indicate the bit rate, type of service, type of error protection coding, etc. (TFI = Traffic Format Indicator).

Fig. 3 shows on the basis of a GSM radio scenario with z. B. two radio cells and base stations arranged therein (base transceiver station), a first base station BTS1 (transmitter / receiver), a first radio cell FZ1 and a second base station BTS2 (transceiver) omnidirectionally "illuminating" a second radio cell FZ2 FDMA / TDMA / CDMA radio scenario, in which the base stations BTS1, BTS2 have an air interface designed for the FDMA / TDMA / CDMA radio scenario with a plurality of mobile stations MS1 located in the radio cells FZ1, FZ2. . . MS5 (transceiver) by wireless unidirectional or bidirectional - upward direction UL (up link) and / or downward direction DL (down link) - telecommunication on corresponding 'transmission channels TRC (transmission channel) are connected or connectable. The base stations BTS1, BTS2 are connected in a known manner (cf. GSM telecommunications system) to a base station controller BSC (BaseStation Controller) which takes over the frequency management and switching functions as part of the control of the base stations. The base station controller BSC is in turn via a mobile switching center MSC (Mobile Switching Center) with the higher-level telecommunications network, for. B. the PSTN (Public Switched Telecommunication Net work) connected. The mobile switching center MSC is the administration center for the telecommunications system shown. It takes over the complete call management and, with associated registers (not shown), the authentication of the telecommunications participants and the location monitoring in the network.

Fig. 4 shows the basic structure of the base station designed as a transceiver BTS1, BTS2, while Fig. 5 shows the basic structure of the mobile station also designed as a transceiver MT1. . . MT5 shows. The base station BTS1, BTS2 takes over the sending and receiving of radio messages from and to the mobile station MTS1. . . MTS5, while the mobile station MT1. . . MT5 takes over the sending and receiving of radio messages from and to the base station BTS1, BTS2. For this purpose, the base station has a transmitting antenna SAN and a receiving antenna EAN, while the mobile station MT1. . . MT5 has a common antenna ANT that can be controlled by an antenna switchover AU for sending and receiving. In the upward direction (receive path), the base station BTS1, BTS2 receives, for example, at least one radio message FN with an FDMA / TDMA / CDMA component from at least one of the mobile stations MT1 via the receive antenna EAN. . . MT5, while the mobile station MT1. . . MT5 in the downward direction (reception path) receives, for example, at least one radio message FN with an FDMA / TDMA / CDMA component from at least one base station BTS1, BTS2 via the common antenna ANT. The radio message FN consists of a broadband dig spread carrier signal with a modulated information composed of data symbols.

In a radio receiving device FEE (receiver) that is received carrier signal filtered and on an intermediate fre  quenz downmixed, which in turn scanned further and quantized. After an analog / digital conversion the signal that is on the radio path through multipath propagation has been distorted, an equalizer EQL, which the To a large extent compensates for distortions (Stw .: Synchro nization).

The KS is then tried in a channel estimator Transmission characteristics of the transmission channel TRC on the the radio message FN has been transmitted. The The transmission properties of the channel are in the time domain richly indicated by the channel impulse response. So that the ka nal impulse response can be estimated, the radio will directs FN on the sending side (in the present case by Mobilsta tion MT1. . . MT5 or the base station BTS1, BTS2) a spec Elle, trained as a training information sequence formation in the form of a so-called Mitambel assigned or assigned.

In a subsequent one for all received signals common data detector DD are those in the common Signal contained individual mobile station-specific Equalized and separated signal components in a known manner. To equalization and separation are Data converter SDW the previously available data symbols in bi converted to binary data. Then in a demodulator DMOD the original bit stream is obtained from the intermediate frequency, before the individual time slots in a demultiplexer DMUX the right logical channels and therefore also the ones below be assigned to different mobile stations.

The bit sequence obtained is decoded channel by channel in a channel codec KC. Depending on the channel, the bit information is assigned to the control and signaling time slot or a voice time slot and - in the case of the base station ( FIG. 4) - the control and signaling data and the voice data for transmission to the base station controller BSC together for signaling and voice coding / - de coding (voice codec) responsible SS give, while - in the case of the mobile station ( Fig. 5) - the control and signaling data of a control and signaling unit STSE responsible for complete signaling and control of the mobile station and the voice data for the voice input and output designed voice codec SPC are passed.

In the speech codec of the interface SS in the base station BTS1, BTS2 the speech data in a predetermined Da current (e.g. 64 kbit / s current in the network direction or 13 kbit / s Current from network direction).

The complete control of the Base station BTS1, BTS2 performed.

The base station transmits in the downward direction (transmission path) BTS1, BTS2 via the SAN transmission antenna, for example least one radio message FN with an FDMA / TDMA / CDMA compo to at least one of the mobile stations MT1. . . MT5, true rend the mobile station MT1. . . MT5 in the upward direction (Transmission path) via the common antenna ANT, for example at least one radio message FN with an FDMA / TDMA / CDMA Component at at least one base station BTS1, BTS2 sen det.

The transmission path begins at the base station BTS1, BTS2 in FIG. 4 with the fact that control and signaling data and voice data received in the channel codec KC from the base station controller BSC via the interface SS are assigned a control and signaling time slot or a voice time slot and they are coded channel by channel into a bit sequence.

The transmission path starts at the mobile station MT1. . . MT5 in Fig. 5 so that in the channel codec KC from the speech codec SPC voice data and from the control and signaling unit STSE control and signaling data are assigned to a control and signaling time slot or a voice time slot and these channel by channel are encoded in a bit sequence.

Those in the base station BTS1, BTS2 and in the mobile station MT1. . . MT5 obtained bit sequence is stored in one data to symbol converter DSW converted into data symbols. In the An this is followed by the data symbols in a spreader equipment SPE with one individual participant Spread code. In the burst generator BG, consisting of a Burst assembler BZS and a multiplexer MUX will be there according to the spread in the burst composer BZS Data symbols a training information sequence in the form of a Mitambel added to the channel estimation and in the multiplexer MUX the burst information obtained in this way onto the always set the correct time slot. Finally, the burst obtained in each case in a modulator MOD high-frequency modulated as well as digitally analog converted before that on signal obtained in this way as a radio message FN via a Radio transmission device FSE (transmitter) on the transmission antenna SAN or the common antenna ANT is emitted.

The fingers of a RAKE receiver can lysed using an ear and late tracking procedure (see JG Proakis:.. "Digital Communications", McGraw-Hill, Inc. 3 ^rd Edition, 1995, Chapter 6.3) without a renewed, time and having to perform resource-intensive channel estimation according to the change in the transmission channel. To do this, two additional fingers are added to each RAKE finger. The two fingers detect the reception signal r (t) with the same spreading code s (t) as the main finger, the only difference to the main finger is that the reception signal is advanced one position for the early finger and one sample position for the late finger ( delay) is delayed. This procedure should be used in particular for oversampling. The energies collected by the early and late fingers are compared. The finger position of the main finger is shifted in the direction of the stronger finger after this comparison. This is only carried out when a certain threshold value TH is exceeded in the energy difference. The RAKE receiver is in the indicated literature (see JG Proakis, "Digital Communications", Chapter 14.5, McGraw-Hill, Inc. 3 ^rd Edition., 1995.) Described in detail.

Fig. 6 shows the principle for a RAKE finger with the corresponding early and late fingers.

From Fig. 6 it can be seen that the early finger performs the despreading process for the received signal by a delay unit earlier than the actual main finger. The late finger performs the de-spreading process exactly one delay later than the main finger.

In Fig. 7, the construction is shown a finger. It consists essentially of two multipliers and an accumulation unit. Each sampled received value r (t) is multiplied by the spreading code s (t) and weighted according to a channel estimate by the weight g, which is different for each finger of a RAKE.

Now the values calculated in this way are added up according to the spreading factor. The result of each finger is a complex signal, which is a despread symbol. In the case of the ear ly and late fingers, the multiplication by the weight can be omitted, ie the weight is 1. All of the signals shown in FIGS. 6 and 7 are complex and thus consist of a real part and an imaginary part.

The evaluation of the results that the early and late Deliver fingers happens through the formation of the amount and a subsequent comparison of the amounts. If the amounts ge are significantly different, that is a minimum have a difference which is determined by a value TH,  the position of the fingers is changed so that the Main finger after the change in position with the greater energy.

This is illustrated in FIG. 8. The energy calculated by the ear ly finger, referred to here as e, is compared to the energy l calculated by the late finger. This is done simply by evaluating the energy difference between the two fingers. In the first case, the fingers are not moved, since the difference between early and late energy is not particularly large, that is to say essentially smaller than the threshold value TH to be defined. In the second case, the difference between the early and late fingers is greater than TH and the energy of the late finger is greater than the energy of the early finger. It follows that the main finger is moved back by one delay stage. In the third case, the difference between early and late fingers is also greater than TH and this time the energy of the early finger is greater than the energy of the late finger. It follows that the main finger is shifted forward by one delay.

Calculating energies and making decisions for the tracking is shown below. With a + ib that is The result of the early finger and c + id is the result of the La te fingers. These results are shown in the Result registers filed.

The following calculations are carried out to decide the tracking direction:

• 1. Calculation of the energy of the early finger e = sqrt (a ² + b ² )
• 2. Calculation of the energy of the late finger l = sqrt (c ² + d ² )
• 3. Calculation of x = e - l
• 4. Evaluation:
  If x <- TH then position your finger one position later.
  If x <TH then position your finger one place earlier.

The standard simplification of this algorithm is achieved by omitting the root calculation (sqrt). Since the root function (here called sqrt) is strictly monotonous, this can be done without restricting the generality. This means that steps 1 and 2 are replaced by the following steps:

• 1. Calculation of the energy of the early finger e = a ² + b ²
• 2. Calculation of the energy of the late finger l. = c ² + d ²

Fig. 9 shows the structure of a hardware that can perform the early-late control.

The multipliers calculate the squares of the inputs a, b, c and d. From this the sums a ² + b ² and c ² + d ^{2 are} calculated. Therefore, four multipliers and two adders are required here. The difference between the two sums is then compared to the threshold value. If the difference is below the negative threshold, this leads to a repositioning of the fingers to the next later position. If the difference is above the positive threshold, this leads to a repositioning of the fingers to the next previous position. The request for repositioning can be read from the outputs. There are three depicted options, repositioning to an earlier position, repositioning to a later position and leaving your finger on the position. In the case of binary coding of the output signals, only these three output values can occur, since simultaneous repositioning to an earlier and a later position is not possible.

The object underlying the invention is a Process for the hardware-based implementation of the finger track ing a rake receiver in telecommunications systems with  wireless telecommunications between mobile and / or sta tional transmitters / receivers, especially in mobile radio systems third generation systems, in which the evaluator tion of the energies received by the rake receiver and the Control the positioning of fingers of the rake receiver are optimized in terms of hardware requirements.

This object is achieved according to the patent claim in that:

• a) main finger of the rake receiver by means of an early Tracking method predetermined early fingers and by means of a late tracking method specified late fingers according to the Change of a transmission channel can be tracked where with a reception signal to the respective main finger preferred for the respective early finger and for the respective late finger is delayed,
• b) that of the respective early finger and the respective Late-fingered energies are compared
• c) the position of the respective main finger depending is maintained by this comparison or in the direction the early finger or the late finger is moved,
• d) the direction in which the position of the respective main finger is shifted depending on whether the relationship
  x = (a - c) (a + c) + (b - d) (b + d),
  where a ² + b ² essentially indicates the energy collected by the early finger, and wherein c ² + d ² indicates the energy collected by the late finger, is smaller or larger than predetermined threshold values.

The idea on which the invention is based consists of the following:
The circuit shown below, which calculates the evaluation of the energies and the control of the positioning of the fingers, is optimized with regard to the hardware requirement. As part of the optimization, it is sufficient to combine the steps 1, 2 and 3 above into one, which performs the following calculation:

• 1. x = (a - c) (a + c) + (b - d) (b + d)
• 2. Evaluation:
  If x <-TH then position finger one position later.
  If x <TH then position your finger one place earlier

The correctness of the solution can be proven as follows:
From x = e - l it follows:

x = (a ² + b ² ) - (c ² + d ² ) = a ² + b ² - c ² - d ² = (a ² - c ² ) + (b ² - d ² )

it follows from the binomial formulas:

x = (a - c) (a + c) + (b - d) (b + d) q. e. d.

Fig. 10 makes the structure of the resulting circuit for deciding on a repositioning significantly. The number of components has remained the same, as has the outer structure of the circuit, but two multipliers could be replaced by an adder and a subtractor.

Comparison of hardware requirements (see Table 1)

The hardware requirements of the two circuits should be on this Place to be compared. On the one hand, the total number of required functional units remained the same. Under the Aspect that the individual functional units internally un but have different hardware requirements here presented solution cheaper compared to the conventional Lö solution.  

Table 1

Hardware requirements

Table 1 shows the number of components required for the two solutions shown. The optimization exists now that multipliers are much more hardware need resources as adders. Especially when in like the speed of an adder through a multi copier to be reached. It follows that the solution with only two multipliers a lower absolute hard demand and consequently a lower energy requirement points.

The effort to implement multipliers as digi tal hardware is in contrast to subtractors and adders enormously. Especially when the speed of a circuit is one important role for the application as it plays for the in Real time running cellular systems is the case for the fast multiplication a lot of hardware Taken purchase. The presented hardware solution reduces the Hardware expenditure enormously optimized by two multipliers could become. There is an additional adder and a additional subtractor necessary. The optimization reflects result in a reduction in space requirements when implementing hardware on a chip again. Furthermore stands the number of gates required directly related to the power requirement of a circuit. The power requirement will this optimization also reduced what modern Cellular devices that are battery operated are very important is the criterion. Overall, a reduction in the Space requirement and the power requirement of the circuit be works.

Claims (1)

Process for the hardware-based implementation of the fingertracking of a rake receiver in telecommunication systems with wireless telecommunication between mobile and / or stationary transceivers, in particular in third generation mobile radio systems, with the following features:

• a) Main fingers of the rake receiver are tracked by means of early fingers specified by an early tracking method and by late fingers specified by a late tracking method according to the change in a transmission channel, where there is a received signal in relation to the respective main finger is preferred to the respective early finger and delayed for the respective late finger,
• b) the energies collected by the respective early finger and the respective late finger are compared,
• c) the position of the respective main finger is maintained depending on this comparison or shifted in the direction of the early finger or the late finger,
• d) the direction in which the position of the respective main finger is shifted depending on whether the relationship
  x = (a - c) (a + c) + (b - d) (b + d),

where a ² + b ² essentially indicates the energy collected by the early finger, and wherein c ² + d ² indicates the energy collected by the late finger, is smaller or larger than predetermined threshold values.