WO2009138946A1

WO2009138946A1 - System and method for de-correlation based active code detection

Info

Publication number: WO2009138946A1
Application number: PCT/IB2009/051937
Authority: WO
Inventors: Xun Fan
Original assignee: Nxp B.V.
Priority date: 2008-05-13
Filing date: 2009-05-12
Publication date: 2009-11-19

Abstract

A receiver for detecting active code channels in a received signal. The receiver includes an active code candidate (ACC) engine, an active code detection (ACD) comparator, and an ACD apparatus. The ACC engine receives a signal from a transmitter and determines a number of candidate active code channels contained in the received signal. The received signal includes a set of candidate active code channels. The ACD comparator compares the number of candidate active code channels to a predetermined size of a spreading factor and determines whether the number of candidates in the candidate active code channel set is less than or equal to the predetermined size of the spreading factor. The ACD apparatus selects an ACD scheme from a plurality of ACD schemes and detects active code channels of the received signal based on the selected ACD scheme.

Description

SYSTEM AND METHOD FOR DE-CORRELATION BASED ACTIVE

CODE DETECTION

Time division- synchronous code division multiple access (TD- SCDMA) is a wireless communication standard based on burst transmissions, where the most basic burst unit is a time slot. Fig. 1 depicts a block diagram of one embodiment of a time slot burst structure 100. At the transmitter, a TD- SCDMA system transmits signals according to a specific structure at each time slot, as shown in Fig. 1. A TD-SCDMA time slot is designed to fit into exactly one burst. The time slot burst structure 100 includes four parts, including a midamble with 144 chips duration, two identical data segments at each side of the midamble with 352 chips duration each, followed by a 16 chips guard period (GP) used to alleviate the effect of multipath delay. A chip is one bit of a direct- sequence spread spectrum code. The chip rate of a code is the number of bits per second (chips per second) at which the code is transmitted (or received).

The performance of active code detection directly affects the performance of joint detection (JD) and is therefore performed prior to JD. Prior to implementing JD on a received signal, the spreading codes used by all code divided users (called active codes, or active code channels) are determined. However, for the receiver of a TD-SCDMA downlink, the user equipment (UE) only knows about the active code channels assigned to it and is not aware of any interference users' code channels. Furthermore, sometime when the traffic rate changes from large to small, there may be no transmission in some of the allocated spreading code channels even for the desired user itself. On each active code channel, the carried data symbol information is spread by a combined spreading code of the channelization code, a channelization code specific multiplier, and a cell scrambling code. The channelization codes used in TD-SCDMA systems are called orthogonal variable spreading factor (OVSF) codes. The midamble of the time slot burst structure 100 includes the channelization codes (i.e., the OVSF codes) related to the midamble codes, which are used for channel estimation at the receiver. TD-SCDMA standardizes three midamble allocation modes: default mode, common mode, and specific mode. Different allocation modes imply different relationships between allocated midambles and OVSF codes.

Conventional matched filter (MF)-based ACD, or M-ACD, is widely used for a single-cell environment (i.e., the inter-cell interference can be neglected). In particular, in a single-cell environment with short multi-path channels (e.g., single-path channels), M-ACD may achieve satisfactory performance with low complexity. However, in the multi-cell case (i.e., the case with strong inter-cell interference), because of the strong correlation between the spreading code channels from intended cells and interference cells, the conventional ACD method suffers severe performance degradation, even in short multi-path channel environments.

Embodiments of a system are described. In one embodiment, the system is a receiver for detecting active code channels in a received signal. The receiver includes an active code candidate (ACC) engine, an active code detection (ACD) comparator, and an ACD apparatus. The ACC engine receives a signal from a transmitter and determines a number of candidate active code channels contained in the received signal. The received signal includes a set of candidate active code channels. The ACD comparator compares the number of candidate active code channels to a predetermined size of a spreading factor and determines whether the number of candidates in the candidate active code channel set is less than the predetermined size of the spreading factor. The ACD apparatus selects an ACD scheme from a plurality of ACD schemes and detects active code channels of the received signal based on the selected ACD scheme. Other embodiments of the system are also described. Embodiments of a method are also described. In one embodiment, the method is a method for active code detection (ACD). The ACD method includes determining a number of candidate active code channels contained in a received signal. The ACD method also includes comparing the number of candidate active code channels to a predetermined size of a spreading factor. The ACD method also includes selecting an ACD scheme from a plurality of ACD schemes based on the number of candidate active code channels in the candidate active code channel set relative to the predetermined size of the spreading factor. The ACD method also includes detecting active code channels from the candidate active code channels based on the selected ACD scheme. Other embodiments of the method are also described

Embodiments of an apparatus are also described. The apparatus includes a joint detection (JD) engine, an active code candidate (ACC) engine, an active code detection (ACD) comparator, and an ACD apparatus. The JD engine generates a filtering output signal based on a filter algorithm applied to a received signal. The ACC engine determines a number of candidate active code channels contained in the received signal. The ACD comparator compares a number of candidate active code channels to a predetermined size of a spreading factor and determines whether the number of candidates in the candidate active code channel set is less than the predetermined size of the spreading factor. The ACD apparatus implements a de-correlation based active code detection (D-ACD) in response to the number of candidate active code channels being less than the predetermined size of the spreading factor. Other embodiments of the apparatus are also described.

Other aspects and advantages of embodiments of the present invention will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, illustrated by way of example of the principles of the invention. Fig. 1 depicts a schematic block diagram of one embodiment of a time slot burst structure.

Fig. 2 depicts a schematic block diagram of one embodiment of a receiver system with an active code detection (ACD) apparatus for the detection of an active code channel. Fig. 3 depicts a schematic block diagram of one embodiment of a channel matrix based on M active code channels.

Fig. 4 depicts a schematic block diagram of one embodiment of an ACD apparatus for the detection of an active code channel.

Fig. 5 depicts an association between midambles and orthogonal variable spreading factor (OVSF) codes.

Fig. 6 depicts a schematic block diagram of one embodiment of a channel matrix based on M' candidate active code channels. Fig. 7 depicts a schematic flow chart diagram of one embodiment of a selective active code detection method for use with the ACD apparatus of Fig. 4.

Fig. 8 depicts a schematic flow chart diagram of one embodiment of the de-correlation-based active code detection operation of the selective code detection method of Fig. 7.

Figs. 9A and 9B depict performance data charts of the de-correlation- based active code detection operation of Fig. 8 relative to the conventional active detection for multi-cell (A) single-path and (B) multi-path environments.

Throughout the description, similar reference numbers may be used to identify similar elements.

Fig. 2 depicts a schematic block diagram of one embodiment of a receiver system 200 with an active code detection (ACD) apparatus 206 for the detection of an active code channel. In one embodiment, the receiver 200 is a component of a user equipment (UE). Alternatively, the receiver 200 is component of a base station (BS). The receiver system 200 includes a received signal sequencer 202, a channel estimator 204, an active code detection (ACD) apparatus 206, and a joint detection (JD) engine 208. Additionally, the receiver system 200 includes at least one bus interface 210, a memory device 214, and a processor 212. In one embodiment, the bus interfaces 210 facilitate communications related to the ACD apparatus 206 and/or active code detection algorithms executing on the receiver 200, including processing active code detection commands, as well as storing, sending, and receiving data packets associated with the active code detection operations of the ACD apparatus 206. Although the depicted receiver 200 is shown and described herein with certain components and functionality, other embodiments of the receiver 200 may be implemented with fewer or more components or with less or more functionality.

In some embodiments, the receiver system 200 receives signals from a transmitter (not shown). Each received signal may include an intended signal and at least one interference signal. In one embodiment, the received signal sequencer 202 sequences the received signal in at least two parts. The received signal sequencer 202 sequences the received signal into a data related received sequence and a midamble related received sequence. In one embodiment, the midamble related received sequence from the received signal sequencer 202 is input to the channel estimator 204 to obtain the estimate of channel state information. Channel state information may include the channel impulse response and a receiver noise variance. The channel estimator 204 outputs a channel state information estimate to the ACD apparatus 206 and the joint detection engine 208.

In one embodiment, the ACD apparatus 206 receives the data related received sequence from the received signal sequencer 202 and the channel state information estimate from the channel estimator 204. Taking the data related received sequence and the channel state information as inputs, the ACD apparatus 206 generates and outputs an estimate of active code channels (i.e., spreading code channels). In some embodiments, at least part of the ACD apparatus 206 208 is incorporated in a processor and/or application-specific integrated circuit (ASIC). Details of the functions and operations of the ACD apparatus 206 are described below or more detail with reference to Fig. 3.

The joint detection engine 208 receives the data related received sequence from the received signal sequencer 202, the channel state information estimate from the channel estimator 204, and the active code channel estimate from the ACD apparatus 206. The joint detection engine 208 then filters the three inputs from the received signal sequencer 202, the channel estimator 204, and the ACD apparatus 206.

The JD engine 208 recovers information contained in the received signal in multiple access interference (MAI) and inter-symbol interference (ISI) channel environments. The input of the JD engine 208 includes the data related received sequence, the channel response (or channel impulse response, CIR) and the active code channels. In some embodiments, JD algorithms include zero-forcing block equalization (ZF-BLE), and minimum mean square error lock equalization (MMSE-BLE) algorithms. For the data detection at each data field, both of the algorithms operate on a derived signal model as follows, which reflects the relationship between the received signals and transmitted information symbols as:

r = Ad + n (1) where r = [r_ι,r₂,...,r_Nβ+w_-_ι]^τ is a NQ + W - 1 -dimensional received signal vector, which is derived from the data related received sequence after midamble interference cancellation. In the TD-SCDMA standard, N = 22 is the number of carried information symbols on an active code channel for the corresponding data field, Q = 16 is the adopted spreading factor, W is the length of a wireless channel in chips, d = [d^(1)r,d^(2)r,...,d^(Λ°^r]^r is an MN- dimensional data vector including information symbols on all the M active code channels for the corresponding data field; d^(B) = \d[ⁿ\d%\...,d™f , n = 1,2,..., JV , is the data vector including the information symbols on all the M active code channels belonging to the same symbol label. n = [n_l, n₂,...,n_NQ+w__lf is the NQ + W - 1 -dimensional additive white Gaussian noise (AWGΝ) vector with the entries modeled as independent and identically distributed (i.i.d.) Gaussian random variables with zero mean and variance N₀ .

In one embodiment, the processor 212 is utilized by the ACD apparatus 206 to perform active code detection operations and related functions.

Alternatively, a separate processor may be coupled to the ACD apparatus 206 and/or other components of the receiver 200. In some embodiments, the processor 212 is a central processing unit (CPU) with one or more processing cores. In other embodiments, the processor 212 is a network processing unit (ΝPU) or another type of processing device such as a general purpose processor, an application specific processor, a multi-core processor, or a microprocessor. In general, the processor 212 executes one or more instructions to provide operational functionality to the receiver 200. The instructions may be stored locally in the processor 212 or in the memory device 214. Alternatively, the instructions may be distributed across one or more devices such as the processor 212, the memory device 214, or another data storage device.

In one embodiment, the memory device 214 is utilized by the ACD apparatus 206 to perform active code detection operations and related functions. Alternatively, a dedicated memory device may be coupled to the ACD apparatus 206 and/or other components of the receiver 200. In some embodiments, the memory device 214 is a random access memory (RAM) or another type of dynamic storage device. In other embodiments, the memory device 214 is a readonly memory (ROM) or another type of static storage device. In other embodiments, the illustrated memory device 214 is representative of both RAM and static storage memory within the receiver system 200. In some embodiments, the memory device 214 is content-addressable memory (CAM). In other embodiments, the memory device 214 is an electronically programmable readonly memory (EPROM) or another type of storage device. Additionally, some embodiments store the instructions related to the operations of the receiver 200 as firmware such as embedded code, basic input/output system (BIOS) code, and/or other similar code.

Fig. 3 depicts a schematic block diagram of one embodiment of a channel matrix 300 based on M active code channels. As depicted in the channel matrix 300, A is a (NQ + W- l)x(MN) equivalent channel matrix where the combined CIR matrix is given as:

V = [b⁽¹⁾,b⁽²⁾,...,b^w] (2)

where the column vector b^(m) = h^(m) *s^(m) , m = \,2,...,M . The column vector is a combined CIR vector formed through the convolution between the CIR vector of the m'^h active code channel h^(m) and its affiliated spreading code vector s^(m) . According to the TD-SCDMA standard, s^(m) is a scrambled OVSF code. In particular,

Im) _ r Jm) Jm) Jm) -,T = [q^(m) * w^(m)

* w^(m) *v₁ ⁽ ₆ ^m)f (3)

where c[^m) , q = 1, 2, ..., 16 , is the q'^h element of the OVSF code vector c^(m)

corresponding to the m'^k active code channel; w^(m) is an OVSF code channel specific multiplier; v_q ^(m) , and q = 1, 2,...,16 , is the q'^h element of the cell scrambling code vector v^(m) =

corresponding to the m^th active code channel. For the JD engine 208 to perform an effective JD operation, h^(m) is obtained by the channel estimator 204, and c^(m) is determined by the ACD apparatus 206 prior to performing JD.

Fig. 4 depicts a schematic block diagram of one embodiment of an ACD apparatus 400 for the detection of an active code channel. In one embodiment, the ACD apparatus 400 is substantially similar to the ACD apparatus 206 of Fig. 2. The ACD apparatus 400 includes an active code candidate engine 402, an ACD comparator 404, an M-ACD engine 406, a D-ACD engine 408, and a data power engine 410. Although the depicted ACD apparatus 400 is shown and described herein with certain components and functionality, other embodiments of the ACD apparatus 400 may be implemented with fewer or more components or with less or more functionality.

In one embodiment, the active code candidate engine 402 determines the number of candidate active code channels included in an active code channel set. Candidate active code channels are determined based on the adopted midamble allocation mode and the active channel window information derived from the output of the channel estimator 204.

For example, consider a downlink multi-cell TD-SCDMA system with the detailed configurations of Table 1. In such a TD-SCDMA system, the intended receiver, UE A, receives a downlink signal from a serving cell, Cell A, while suffering the co-channel interference from a neighboring cell, Cell B. For both Cell A and Cell B, the number of users, K_CELL, is eight and the midamble allocation mode adopted is the default mode. Thus, according to the active channel window information obtained from the output of the channel estimator 204, which indicates the midamble shift information m^ and m^ , UE A determines a candidate active code channel set as:

(4)

which includes four possible active code channels respectively corresponding to four possible OVSF codes used for the transmission from the two cells (i.e., the OVSF codes c?) and c?J for the transmission from Cell A, and c?) and c?J for the transmission from Cell B). Among all of the four OVSF codes, the primary OVSF codes associated with the corresponding midamble (i.e., the OVSF codes C_jg for both cells) are determined to be used by the receiver, while the secondary OVSF codes associated with the corresponding midamble (i.e., the OVSF codes C_jg ^} for both cells) are unknown whether they are used, and include the codes to be detected by the ACD apparatus 206. An example configuration is shown in Table 1.

Table 1. S stem confi urations for a downlink multi-cell TD-SCDMA s stem

From the example configuration of Table 1 , the information transmitted from Cell B to User B leads to co-channel interference (or MAI) at the receiver of User A, which is handled by the JD engine 208. The midamble mode used is the default midamble for both Cell A and Cell B. Fig. 5 depicts an association between midambles and orthogonal variable spreading factor (OVSF) codes for Kc_ELL ⁼ 8. Thus, according to the association between midambles and OVSF codes depicted in Fig. 5, m^ and m^ are adopted in the transmission from Cell A and Cell B, respectively. For m^' , the subscript n indicates the index of the basic midamble used in the corresponding cell, and the superscript k indicates the index of the midamble shift. In one embodiment, an OVSF code of SF = 16 is configured for use in the downlink transmission.

In one embodiment, the ACD comparator 404 compares the number of active code channel candidates, as determined by the active code candidate engine 402, to the adopted spreading factor, Q. In some embodiments, such as the TD- SCDMA example, Q = 16. Alternatively, the adopted spreading factor is another value. Thus, in the TD-SCDMA example, the active code candidate engine 402 determines the number of active code channel candidates and the ACD comparator 404 compares the number of active code channel candidates to the predetermined spreading factor, which is Q = 16 in the TD-SCDMA example. In some embodiments, the ACD comparator 404 compares a filtering output data power of an unknown active code channel to a predetermined threshold.

In one embodiment, the M-ACD engine 406 performs a matched filter active code detection scheme to detect unknown active code channels in the received signal. In some embodiments, the M-ACD engine 406 performs a matched filter active code detection scheme in response to the number of candidate active code channels being greater than or equal to the adopted spreading factor, Q. Thus, in the TD-SCDMA example, the active code candidate engine 402 determines the number of active code channel candidates and the ACD comparator 404 compares the number of candidate active code channels to the predetermined spreading factor, Q = 16. When the ACD apparatus 400 determines that the number of candidate active code channels is greater than or equal to 16, then the ACD apparatus 400 implements the M-ACD engine 406 to perform the M-ACD procedure and, thus, the ACD apparatus 400 selects the M- ACD engine 406 to compute the active code channels based on the number of candidate active code channels in the candidate active code channel set. Letting r = [r₁,r₂,...,r_Nβ+w_₁]^τbc a (NQ + W- l)-dimensional received signal vector as described above in Eq. (1), the M-ACD engine 406 performs the matching filter on the signal vector as follows:

5 d = A'"r (5)

where A' is a (NQ + W-V)x(M'N) channel matrix depicted in Fig.6. Fig.6 depicts a schematic block diagram of one embodiment of a channel matrix based on M' candidate active code channels. The channel matrix of Fig.6 has the same 10 structure as that in Fig.3 except that all of the OVSF codes in the candidate active code channel, M', are included, where:

V' = [b'⁽¹⁾,b'⁽²⁾,b'⁽²⁾,b'⁽⁴⁾] (6)

15 is the combined CIR matrix with the column vector b'^(m) =h'^(m)*s'^(m),m= l,2, ..., 4, which is a combined CIR vector formed through the convolution between the CIR vector of the m'^h candidate code channel h'^(m) and its affiliated spreading code vector s ^t(m) , given by:

20 s'⁽¹⁾ -IV⁽¹⁾ v'⁽¹⁾ v'⁽¹⁾f

- - lI>C₁ ⁽¹⁾ * w W;⁽¹⁾ * v V₁ ⁽⁴⁾ , rC₂ ⁽¹⁾ * w W;⁽¹⁾ * v V₂ ⁽⁴⁾ ,-, rC₁ ⁽ ₆ ¹⁾ * w W;⁽¹⁾ * v V₁ ⁽ ₆ ⁴⁾ f J s.(2) _ r ,(2) ,(2) ,(2)-,r

= [_Cl ⁽²⁾ *w⁽²⁾ *v₁ ⁽⁴⁾,^²⁾ *w⁽²⁾ *v₂ ⁽⁴⁾,...,c₁ ⁽ ₆ ²⁾*w⁽²⁾ *v₁ ⁽ ₆ ⁴⁾f βt(3) _ r_eι(3) .(3) _e'(3)ηr

a ?S - - fIrC₁ ⁽¹⁾ * M w;⁽¹⁾ * v V₁ ⁽⁸⁾ , rc₂ ⁽¹⁾ * M w;⁽¹⁾ * v V₂ ⁽⁸⁾ ,..., rC₁ ⁽ ₆ ¹⁾ * w w;⁽¹⁾ * v V₁ ⁽ ₆ ⁸⁾ l j^r βl(4) _ r t(4) ι(4) _e'(⁴)n^r

- - T Lcc₁ ⁽²⁾ * w w⁽²⁾ * v V₁ ⁽⁸⁾ , cc₂ ⁽²⁾ * w w⁽²⁾ * v V₂ ⁽⁸⁾ ,..., cc₁ ⁽ ₆ ²⁾ * w w⁽²⁾ * v V₁ ⁽ ₆ ⁸⁾ f j f \ 7/) ) where w^m and w^{2) are the standard-defined multipliers for the OVSF codes cj_g and cj^ , respectively, and the other variables are as explained in Table 1. d = [d^(1)r,d^(2)r,...,d^(W)r]^r is a MTV-dimensional filtering output vector where d^(B) = [d[ⁿ⁾ ,d™ ,..., d%}f , n = 1,2,...,N , is the vector including the filtering output for all the M' candidate active code channels belonging to the same symbol label in the concerned data field.

Due to the particular structure of A ' , the match filtering in Eq. (5) is performed symbol by symbol by the M-ACD engine 406 as follows:

a^(κ) = V'^// r_(κ__{1)ρ+l κρ+}^₁ , n = l,2,...,N (8)

where r_m „ denotes the sub-vector formed by taking the m^th to n^th entries of vector r .

In one embodiment, the D-ACD engine 408 generates a symbol-level de- correlation filtering matrix and performs a de-correlation-based active code detection algorithm according to the generated symbol-level de-correlation filtering matrix. In other words, the D-ACD engine 408 filters received signals through the symbol-level de-correlation filtering matrices symbol by symbol. In some embodiments, the D-ACD engine 408 stores the computed symbol-level de- correlation filtering matrix in the memory device 214. In some embodiments, the D-ACD engine 408 performs a de-correlation-based active code detection scheme in response to the number of candidate active code channels. In some embodiments, the D-ACD engine 408 performs de-correlation-based active code detection when the number of candidate active code channels is less than the adopted spreading factor, Q. Thus, in the TD-SCDMA example of Table 1, the active code candidate engine 402 determines the number of active code channel candidates, and the ACD comparator 404 compares the number of active code channel candidates to the predetermined spreading factor, Q = 16. When the ACD apparatus 400 determines that the number of candidate active code channels is less than or equal to 16, then the ACD apparatus 400 implements the D-ACD engine 408 to perform de-correlation-based active code detection. Thus, in some embodiments, the ACD apparatus 400 selects the D-ACD engine 408 to compute the active code channels based on a determination of the number of candidate active code channels in the candidate active code channel set. When the number of code channels in the candidate active code channel set is not greater than the spreading factor, the D-ACD engine 408 performs symbol-level de-correlation filtering on the received signals as follows:

" ^~~ *^Ja'ecorr^r(»-l)g+l nQ+W-1

- CV'^H VT¹V⁸ T n - \ 2 N (9)

where d^(B) = [d[ⁿ⁾ , d₂ ⁽ⁿ⁾ , ... , d%] f , n = 1, 2, ... , N , is the vector including the filtering output for all the M' candidate active code channels belonging to the same symbol label in the concerned data field; G_decorr = (Y '^H V ')^"1 Y '^H is the symbol-level de- correlation filtering matrix. The elements in the filtering output d are divided into M' groups, each of which is associated with an OVSF code as follows:

7 uj^(m) - — \ Lal₁ ^{m) ,a l₂ ^{m) ,..., la_N ^(m) Λ J^τ

= [d∞, d∞,..j™ ]^τ , m = l, 2,...,M ' (10)

In one embodiment, the data power engine 410 calculates the filtering output data power associated with each unknown active code channel contained in the candidate active code channel set. In some embodiments, the filtering output data power for each OVSF code in the candidate active code channel set is calculated by the data power engine 410 according to:

∑K^(m) |² , m = \,2,...,M' (11)

In some embodiments, the ACD comparator 404 compares the filtering output data power for each OVSF code in the candidate active code channel set to a predefined data power threshold to determine whether the code channel is active. In some embodiments, the comparison threshold is selected according to the filtering output data power for known active code channels. Alternatively, the comparison threshold is selected according to a noise power.

In particular, from the example configuration in Table 1 , consider the active code channel determination according to the filtering output data power for known active code channels. In the example, the known active code channels are the code channel ac and ac in the set of candidate active code channels, i.e., the cfs code channel for Cell A (corresponding to midamble m^^{1 )} ) and the cf_g code for Cell B (corresponding to midamble m^ ), respectively. The unknown active code channels are the code channel αc⁽²⁾ and αc⁽⁴⁾ in the set of candidate active code channels, i.e., the C₁^ code channel for Cell A (corresponding to midamble m^^{1 )} ) and the cj^ code for Cell B (corresponding to midamble m^ ), respectively. Thus, in some embodiments, the ACD comparator 404 compares the data power of an unknown active code channel with that of a corresponding known active code channel with the same midamble. From the comparison, a determination of the active characteristics of the unknown active code channel is made. If the data power of the unknown active code channel is greater than or equal to a multiple of that of the corresponding known active code channel, the unknown active code channel is determined to be active; otherwise, it is determined to be inactive, as follows: a_c(2) ₌ I a"c—tive P - ^{2)

inactive P⁽²⁾ < T/¹

(12) a_cW [active P^ > T_gP^ [ inactive P⁽⁴⁾ < T_gP^(y)

where T^ is a pre-defined relative threshold.

Recalling that the above description is aiming at the filtering on a certain data field out of the two data fields in a timeslot, and that both data fields have the same active code channel characteristics, in some embodiments, the output data power for each active code channel on the two data fields is combined (i.e., added). In one embodiment, the data power for each active code channel on the two data fields is combined to generate a resulting data power for the corresponding active code channel.

Comparing Eq. (8) with Eq. (9), it can be seen that the G₁^_0n. -based filtering herein can remove the interference among the M' candidate active code channels on a symbol period while the traditional MF cannot. This is the main reason for the performance improvement of the proposed D-ACD over the traditional MF -based one, as can be seen in Figs. 9A and 9B.

Table 2. Complexity analysis of D-ACD and conventional M-ACD for the filterin on a s mbol eriod unit: number of com lex multi lication

Table 2 demonstrates that for the filtering of a symbol period, the additional complexity of M ^a (Q + W ) + M ^β is present in D-ACD over M-ACD.

However, the additional complexity is generally acceptable in light of its comparison to the complexity of JD, which exceeds that of D-ACD. JD generally accounts for the majority of the complexity at the receiver 200.

Fig. 7 depicts a schematic flow chart diagram of one embodiment of a selective code detection method 700 for use with the ACD apparatus 400 of Fig.

4. Although the de-correlation-based active code detection method 700 is described in conjunction with the ACD apparatus 400 of Fig. 4, some embodiments of the method 700 may be implemented with other types of ACD apparatuses.

At block 702, the active code candidate engine 402 determines the number of available active code channel candidates in a received signal. The number of candidate active code channels in a received signal make up the candidate active code channel set. At block 704, the ACD comparator 404 compares the number of candidates to the size of the spreading factor. The spreading factor, Q, implemented in the TD-SCDMA standard is 16. In one embodiment, the spreading factor is an orthogonal variable spreading factor (OVSF). OVSF is an implementation of code division multiple access (CDMA), including TD- SCDMA, where before transmission of each signal, the spectrum is spread through the use of a user's code. Code channels' codes are chosen to be mutually orthogonal to each other. These codes are derived from an OVSF code tree, as depicted in Fig. 5, and each code channel is given a different, unique code.

At block 706, the ACD comparator 400 determines whether the number of candidates in the candidate active code channel set is less than or equal to the adopted spreading factor. The de-correlation is valid under the condition that the dimension of the signal space to be de-correlated is not greater than that of observation space. Subsequently, for the symbol-level de-correlation filtering to be performed, this condition implies that M' ≤ Q + W-l for the Q + W-lxM' symbol-level equivalent channel matrix V ' (as shown in Fig. 6). Thus, if the number of code channels in the candidate active code channel set (i.e., M ') is not greater than the spreading factor (i.e., Q), the symbol-level de-correlation (i.e., the matrix V ' based de-correlation) can be effectively used in various channel environments, including single-mode and multi-mode environments. When the candidates are determined to be greater than the adopted spreading factor, at block 708, the M-ACD engine 406 performs the matched filter (MF)-based active code detection. Otherwise, at block 710, the D-ACD engine 408 performs de-correlation-based active code detection based on the symbol- level de-correlation filtering matrix, Gdeco_rr- At block 712, the data power engine 410 calculates the filtering output data power for each candidate active code channel. At block 714, the ACD apparatus 400 selects one of the unknown candidates in the candidate active code channel set in order to compare the filtering output data power of the selected candidate to a predetermined threshold. In some embodiments, the comparison threshold is selected according to the filtering output data power for known active code channels. Alternatively, the comparison threshold is selected according to a noise power. At block 716, the ACD comparator 404 compares the filtering output data power of a selected candidate with the predetermined threshold to determine whether the filtering output data power of the selected candidate is greater than or equal to the predetermined threshold.

When the ACD comparator 404 determines that the filtering output data power of the selected candidate is less than the predetermined threshold, then the ACD apparatus 400 designates the selected candidate as an inactive code channel and the ACD apparatus 400 selects another unknown candidate from the candidate active code channel set at block 714. Otherwise, at block 720, the ACD apparatus 400 designates the selected candidate as an active code channel and the ACD apparatus 400 selects another unknown candidate from the candidate active code channel set at block 714.

Fig. 8 depicts a schematic flow chart diagram of one embodiment of the de-correlation-based active code detection operation 710 of the selective code detection method 700 of Fig. 7. Although the de-correlation-based active code detection operation 710 is described in conjunction with the ACD apparatus 400 of Fig. 4, some embodiments of the operation 710 may be implemented with other types of ACD apparatuses.

At block 802, the D-ACD engine 408 computes the symbol-level de- correlation filtering matrix, G_de∞rr, defined by G_decorr = (V '^H V ')^ V '^H . In some embodiments, the D-ACD engine 408 stores the symbol-level de-correlation filtering matrix in the memory device 214. At block 804, the D-ACD engine 408 computes the symbol-level de-correlation algorithm based on the computed symbol-level de-correlation filtering matrix.

Thus, for the case that the number of candidate active code channels is not greater than spreading factor, the proposed D-ACD performs the symbol-level de- correlation filtering instead of the traditional matched filtering on the received signals to obtain the data power for each candidate active code channel. The D- ACD engine 408 then determines the active characteristics for each unknown active code channel through the comparison of its data power with a threshold. Figs. 9A and 9B depict performance data charts 900 and 902 of the de- correlation-based active code detection method 700 relative to the conventional active code detection for multi-cell (A) single-path and (B) multi-path environments.

Performance data chart 900 illustrates simulation results to compare the performance of D-ACD 906 to the conventional M-ACD 904 in a two-cell TD- SCDMA system (as described in Table 1) under a single-path channel environment. As can be seen in Fig. 9A, in an environment with strong inter-cell interference, the performance of the conventional M-ACD 904 generally tapers off relative to the signal-to-noise ratio (SNR). In other words, the probability of detecting an error with the conventional M-ACD 904 remains relatively the same as SNR increases. On the other hand, performance of the D-ACD 906 in the single-path environment continually increases its performance advantage over the performance of the conventional M-ACD 904 as the SNR increases.

Performance data chart 902 illustrates simulation results to compare the performance of D-ACD 910 to the conventional M-ACD 908 in a two-cell TD- SCDMA system (as described in Table 1) under a multi-path channel environment. As can be seen in Fig. 9B, in an environment with strong inter-cell interference, the performance of the conventional M-ACD 908 improves slightly with the SNR. In other words, the probability of detecting an error with the conventional M-ACD 904 improves slightly as SNR increases. On the other hand, performance of the D-ACD 906 in the multi-path environment continually increases its performance advantage over the performance of the conventional M- ACD 904 as the SNR increases.

Unlike the conventional active code detection scheme in which an M-ACD scheme is implemented across all ranges of candidate active code channels regardless of the adopted spreading factor, embodiments of the D-ACD scheme filter received signals through a symbol-level de-correlation filtering matrix instead of the matched filter for the case that the number of candidate active code channels is less than or equal to the adopted spreading factor. Embodiments of D- ACD achieve significant performance advantage over M-ACD in both single-path and multi-path channel environments for the case of having fewer number of candidate active code channels in relation to the spreading factor.

Embodiments of the D-ACD method 700 described with reference to Fig. 7 and illustrated with reference to Figs. 9A and 9B have a demonstrable impact on the efficiency of active code detection in downlink receivers. Embodiments of the D-ACD method 700 effectively reduce the probability of errors in detecting unknown active code channels in a received signal with relatively little increase in complexity when compared to joint detection (JD) complexity. Moreover, embodiments of the described D-ACD method 700 provide varying tradeoffs between performance and complexity for practical receiver implementations.

Although the operations of the method(s) herein are shown and described in a particular order, the order of the operations of each method may be altered so that certain operations may be performed in an inverse order or so that certain operations may be performed, at least in part, concurrently with other operations. In another embodiment, instructions or sub-operations of distinct operations may be implemented in an intermittent and/or alternating manner.

Although specific embodiments of the invention have been described and illustrated, the invention is not to be limited to the specific forms or arrangements of parts so described and illustrated. The scope of the invention is to be defined by the claims appended hereto and their equivalents.

Claims

What is Claimed is:

1. A receiver comprising: an active code candidate (ACC) engine to receive a signal from a transmitter and to determine a number of candidate active code channels contained in the received signal, wherein the received signal comprises a set of candidate active code channels; an active code detection (ACD) comparator coupled to the ACC engine, the ACD comparator to compare the number of candidate active code channels to a predetermined size of a spreading factor and to determine whether the number of candidates in the candidate active code channel set is less than or equal to the predetermined size of the spreading factor; and an ACD apparatus coupled to the ACC engine, the ACD apparatus to select an ACD scheme from a plurality of ACD schemes and to detect active code channels of the received signal based on the selected ACD scheme.

2. The receiver of claim 1, wherein the ACD apparatus is further configured to select a de-correlation-based ACD (D-ACD) scheme from the plurality of ACD schemes based on the number of the candidate active code channels being less than or equal to the predetermined size of the spreading factor and to implement the D-ACD scheme to detect the active code channels of the received signal.

3. The receiver of claim 2, further comprising a D-ACD engine coupled to the ACD apparatus, the D-ACD engine to generate a symbol-level de-correlation filtering matrix in response to the selection of the D-ACD scheme by the ACD apparatus, wherein the D-ACD engine filters the received signal through the symbol-level de-correlation filtering matrix.

4. The receiver of claim 3, further comprising a memory storage device coupled to the D-ACD engine, the memory storage device to store the generated symbol-level de-correlation filtering matrix.

5. The receiver of claim 3, wherein the D-ACD engine is further configured to perform the D-ACD scheme on the candidate active code channels according to the generated symbol level de-correlation filtering matrix in response to the selection of the D-ACD scheme by the ACD apparatus, wherein the D-ACD scheme is defined by:

- (V '^H V¹V¹V ¹^ r 77 - 1 ? N

where G_decorr = (V ^rH V ')^ V ^rH is the symbol-level de-correlation filtering matrix,

V is a combined channel impulse response matrix, N is a number of carried information symbols on an active code channel for a corresponding data segment, Q is the predetermined spreading factor, and W is a length of a wireless channel in chips.

6. The receiver of claim 1, wherein the ACD apparatus is further configured to select a matched- filter-based ACD (M-ACD) scheme from the plurality of ACD schemes based on a number of candidate active code channels in the candidate active code channel set being greater than the predetermined size of the spreading factor and to implement the M-ACD scheme to detect the active code channels of the received signal.

7. The receiver of claim 6, wherein the M-ACD engine is further configured to perform the M-ACD scheme on the candidate active code channels in the candidate active code channel set in response to the selection of the M-ACD scheme by the ACD apparatus, wherein the M-ACD scheme is defined by:

^{U ~ V r}(κ-l)β+l κβ+W-1 > " ^~ -1, ^, -", JV

where r_m „ denotes a sub-vector formed by taking an m^th to n^th entries of vector r,

V is a combined channel impulse response matrix, Λ^ is a number of carried information symbols on an active code channel for a corresponding data segment, Q is the predetermined spreading factor, and W is a length of a wireless channel in chips.

8. The receiver of claim 1, further comprising a joint detection (JD) engine coupled to the ACD apparatus, the joint detection engine to generate a filtering output signal based on a filter algorithm applied to the received signal, and to perform joint detection on the received signal according to the ACD scheme selected and implemented by the ACD apparatus.

9. An active code detection (ACD) method, comprising: determining a number of candidate active code channels contained in a received signal; comparing the number of candidate active code channels to a predetermined size of a spreading factor; selecting an ACD scheme from a plurality of ACD schemes based on the number of candidate active code channels in the candidate active code channel set relative to the predetermined size of the spreading factor; and detecting active code channels from the candidate active code channels based on the selected ACD scheme.

10. The ACD method of claim 9, further comprising: selecting a de-correlation-based ACD (D-ACD) scheme from the plurality of ACD schemes in response to the number of candidate active code channels being less than or equal to the predetermined size of the spreading factor; generating a symbol-level de-correlation based filtering matrix; filtering the received signal through the symbol-level de-correlation filtering matrix; and performing the D-ACD scheme based on the symbol-level de-correlation based filtering matrix in order to detect the active code channels of the received signal.

11. The ACD method of claim 9, further comprising: selecting a matched filter based ACD (M-ACD) scheme from the plurality of ACD schemes in response to the number of candidate active code channels being greater than the predetermined size of the spreading factor; generating a matched filter; filtering the received signal through the matched filter; and performing the M-ACD scheme based on the generated matched filter to detect the active code channels of the received signal.

12. The ACD method of claim 9, further comprising: calculating a filtering output data power for each data segment associated with a time burst in the received signal; comparing the calculated filtering output data power of each data segment to a predetermined power threshold; designating a candidate active code channel as an inactive code channel in response to the calculated filtering output data power of the candidate active code channel being less than the predetermined power threshold.

13. The ACD method of claim 9, further comprising: calculating a filtering output data power for each data segment associated with a time burst in the received signal, wherein each time burst comprises a first data segment allocated to one active code channel and a second data segment allocated to the same one active code channel; comparing the calculated filtering output data power of each data segment to a predetermined power threshold; designating a candidate active code channel as an active code channel in response to the calculated filtering output data power of the candidate active code channel being greater than or equal to the predetermined power threshold.

14. The ACD method of claim 13, further comprising combining the calculated filtering output data power of the first data segment with the calculated filtering output data power of the second data segment to generate a combined data power for the active code channel allocated to the first and second data segments, wherein the combined data power provides an average of the calculated filtering output data power associated with the first and second data segments.

15. An apparatus, comprising: a joint detection (JD) engine to generate a filtering output signal based on a filter algorithm applied to a received signal; an active code candidate (ACC) engine coupled to the JD engine, the ACC engine to determine a number of candidate active code channels contained in the received signal; an active code detection (ACD) comparator coupled to the ACC engine, the ACD comparator to compare a number of candidate active code channels to a predetermined size of a spreading factor and to determine whether the number of candidates in the candidate active code channel set is less than or equal to the predetermined size of the spreading factor; and an ACD apparatus coupled to the ACC engine, the ACD apparatus to implement a de-correlation based active code detection (D-ACD) in response to the number of candidate active code channels being less than or equal to the predetermined size of the spreading factor.

16. The apparatus of claim 15, wherein the ACD apparatus further comprises a D-ACD engine, the D-ACD engine to generate a symbol-level de-correlation based filtering matrix, and to filter the received signal through the symbol-level de-correlation filtering matrix.

17. The apparatus of claim 16, wherein the ACD apparatus is further configured to perform the D-ACD scheme based on the symbol-level de- correlation based filtering matrix in order to detect active code channels associated with the received signal.

18. The apparatus of claim 15, wherein the ACD apparatus further comprises a data power engine, the data power engine to calculate a filtering output data power for each data segment associated with a time burst in the received signal, wherein each time burst comprises a first data segment allocated with one active code channel and a second data segment allocated with the same one active code channel, wherein the ACD comparator is further configured to compare the calculated filtering output data power of each data segment to the predetermined power threshold, and wherein the ACD apparatus is further configured to designate a candidate active code channel as an active code channel in response to the calculated filtering output data power of the candidate active code channel being greater than or equal to the predetermined power threshold.

19. The apparatus of claim 18, wherein the data power engine is further configured to combine the calculated filtering output data power of the first data segment with the calculated filtering output data power of the second data segment to generate a combined data power for the active code channel allocated to the first and second data segments, wherein a combined data power improves an error probability of the selected ACD scheme.

20. The apparatus of claim 15, wherein the ACD apparatus further comprises a data power engine to calculate a filtering output data power for each data segment associated with a time burst in the received signal, wherein the ACD comparator is further configured to compare the calculated filtering output data power of each data segment to the predetermined power threshold, and wherein the ACD apparatus is further configured to designate a candidate active code channel as an inactive code channel in response to the calculated filtering output data power of the one candidate active code channel being less than the predetermined power threshold.