The present invention relates to wireless communication systems. More particularly, the present invention is related to frequency domain joint detection for wireless communication systems, such as multicarrier code division multiple access (MC-CDMA) systems and spread orthogonal frequency division multiple access (OFDMA) systems.
MC-CDMA systems and spread OFDMA systems are candidate systems for third generation partnership project (3GPP) long term evolution that can support a very high data rate for both downlink and uplink transmissions.
MC-CDMA systems are direct spreading code division multiple access (DS-CDMA) systems wherein the spreading is performed in a frequency domain. At the transmitter, N data symbols of each user generated in each OFDM symbol duration are spread into NM chips using a length-M spreading code and mapped to multiple subcarriers. The chips from different users are then added together and the resulting chips are fed to an inverse Fourier transform processor of an OFDM transmitter. The data symbols of each user can be recovered by despreading the output of a Fourier transform processor in a receiver.
OFDMA is an alternative multiple access scheme for OFDM systems, where the signals of different users are separated in the frequency domain by allocating different sub-carriers to different users.
When MC-CDMA or spread OFDMA is used for downlink transmissions, one tap equalizer with despreading across subcarriers may be sufficient for data detection when used subcarriers have the same or very similar channel responses because all the users have the same propagation channel and spreading codes are orthogonal across different subcarriers. However, the subcarriers often have different channel responses, which destroys the orthogonality of codes in the despreading processes, and one-tap matched filter or equalizer may not be sufficient for data detection. When MC-CDMA or spread OFDMA is used for uplink transmissions, one tap equalizer suffers significant performance degradation because users have distinct propagation channels in the uplink and each user experiences different delay spread and fading channel and, therefore, the orthogonality between spreading codes is destroyed.
The present invention provides a method and apparatus for frequency domain joint detection. A receiver considers a distinct delay spread and fading channels for all users and jointly equalizes the channel distortions and detects the data symbols of all users in the frequency domain. The users may be assigned all or a subset of subcarriers with variable spreading factor codes. Alternatively, the subset of subcarriers may be further divided into multiple partitions and the frequency domain joint detection may be performed on the subcarriers in the partition. The users are assigned a spreading code, which may be a Hadamard code, a spread complex quadratic sequence (SCQS) code, generalized chirp-like (GCL) sequences, Newman phase code, polyphase code, or any type of orthogonal or good correlation spreading code. By using the frequency domain joint detection method of the present invention for MC-CDMA or spread OFDMA systems in the uplink or downlink, the performance of data detection is improved when compared to conventional systems.
The features of the present invention may be incorporated into an integrated circuit (IC) or be configured in a circuit comprising a multitude of interconnecting components.
As shown in
Still referring to
Frequency domain joint detection in accordance with a first embodiment of the present invention is described hereinafter. It is assumed that there are N subcarriers in the system, (i.e., a MC-CDMA or spread OFDMA system). The signal model after the FFT operation at the receiver 150 is expressed as follows:
where rn, n=1, 2, . . . , N is the received data in the frequency domain, (i.e., the output of the FFT processor 156 of the receiver 150), for the nth subcarrier. cn(k), k=1, 2, . . . , N is an nth element of the spreading code k. Hn(k) is a flat fading coefficient of the spreading code k in frequency domain for the subcarrier n. d(k) is the transmitted data symbol carried by the spreading code k. Without loss of generality, it is assumed that the data are spread using spreading codes across subcarriers 1 to N. It should be noted that a spreading factor other than N may be used.
Users may use one or several spreading codes for transmissions. Without loss of generality, it is assumed that each user uses only one spreading code and that there are a total of K users in the system. The signal model and detection algorithm can be easily extended to multi-code transmission where one user transmits multiple codes. Equation (1) is generalized to K users and N subcarriers as follows:
where K is the number of users or codes. K and N should satisfy the condition K≦N.
Equation (2) can be written in a vector-matrix form as follows:
{right arrow over (r)}=Af{right arrow over (d)}+{right arrow over (n)}; Equation (3)
where Af is a frequency domain system matrix with a dimension of N×K.
The frequency domain joint detection-performed by the frequency domain joint detector 160 using minimum mean square error (MMSE) criteria is expressed as follows:
{right arrow over ({circumflex over (d)}=(AfHAf+σ2I)−1AfH{right arrow over (r)}; Equation (4)
where {right arrow over ({circumflex over (d)} is the estimated data symbols for all K users, which is output of the frequency domain joint detector 160.
The receiver 150 in
In accordance with a second embodiment of the present invention, variable spreading factors are used for spreading the input data symbols and the frequency domain joint detection is performed on a subset of subcarriers. The smaller the subset of subcarriers on which the frequency domain joint detection is performed the less complex the detection is. It is assumed that there are K total users in the system and each user is assigned a subset of the subcarriers using OFDMA. It is also assumed that there are M subsets of subcarriers, (M≦K), and the i-th subset contains Ni subcarriers. Users may be assigned to the same subset of subcarriers.
Let K1 users (or codes) share the subcarriers of the first subset and use the spreading factor Q1; K2 users (or codes) share the subcarriers of the second subset and use the spreading factor Q2; Ki users (or codes) share the subcarriers of the i-th subset and use the spreading factor Qi. The sum of the number of users (or codes) in all subsets is equal to the total number of users (or codes) in the system, (i.e.,
The received data model for the first subset of subcarriers is expressed as follows:
In general, for the i-th subset of subcarriers, the received signal model can be written as follows:
The spreading factor Qi is equal to the number of subcarriers in the i-th subset, (i.e., Qi=Ni). The overall received signal model can be expressed as follows:
{right arrow over (r)}i=Afi{right arrow over (d)}i+{right arrow over (n)}i,i=1,2, . . . , M. Equation (7)
Where {right arrow over (r)}i is the frequency domain received signal after the FFT processing for the i-th subset of subcarriers. Afi and {right arrow over (d)}i are the frequency domain system matrix and data vector for the i-th subset of subcarriers. The system matrix Afi has a dimension of Qi×Ki. The MMSE frequency domain joint detection performed by the frequency domain joint detector 160 is expressed as follows:
{right arrow over ({circumflex over (d)}i=(Afi
where {right arrow over ({circumflex over (d)}i is the estimated data symbol of the transmitted data symbol {right arrow over (d)}i for the user i. The correlation matrix Afi
In accordance with a third embodiment of the present invention, subcarriers are partitioned and frequency domain joint detection is performed on the subcarriers in the partition for further reduction in complexity. It is assumed that there are M subsets of subcarriers and Di partitions in the i-th subset of subcarriers. There are a total of D partitions,
It is also assumed that there are Ni subcarriers for the i-th subset. Let Ki users use the Di,j partitions and use the spreading factor Qi,j.
The received signal model for the j-th partition of the i-th subset of subcarriers is written as follows:
for the i-th subset. The overall received signal model for the i-th subset and the j-th partition is expressed as follows:
{right arrow over (r)}i,j=Afi,j{right arrow over (d)}i,j+{right arrow over (n)}i,j,i=1,2, . . . , M,j=1,2, . . . , Di; Equation (10)
where {right arrow over (r)}i,j is the received frequency domain signal for the i-th subset and the j-th partition of subcarriers. Afi,j and {right arrow over (d)}i,j are the frequency domain system matrix and data vector for the i-th subset and j-th partition of subcarriers. The system matrix Afi,j has a dimension of Qi,j×Ki. The MMSE frequency domain joint detection performed by the frequency domain joint detector 160 is expressed as follows:
{right arrow over ({circumflex over (d)}i,j=(Afi,j
where {right arrow over ({circumflex over (d)}i,j is the estimated data symbol of the transmitted data symbol {right arrow over (d)}i,j for the i-th subset and j-th partition of subcarriers. The correlation matrix Afi,j
In accordance with a fourth embodiment of the present invention, variable spreading factors are used for users (or codes). Suppose that there are Ki codes in subcarrier subset i and each code is transmitted with a spreading factor Qi,k for the k-th code in subset i. Let Qmax be the maximum spreading factor among spreading factors of all codes in subset i, such that Qmax=max {Qi,1, Qi,2, . . . , Qi,K
Afi=└Af(i,1)A(i,2) . . . Af(i,k) . . . Af(i,K
Each transmission matrix Af(i,k) consists of Qmax rows and Qmax/Qi,k columns. The matrix Af(i,k) is a vector diagonal matrix and can be expressed as:
where
is a column vector which serves as the base or non-zero elements for the j-th column in matrix Af(i,k) containing the code elements and the corresponding channel responses. The p-th element of {right arrow over (v)}(k,j) is:
{right arrow over (v)}(k,j)(p)=cp(k)Hp+(j−1)Q
where Hp(k) is the frequency domain channel response of the p-th subcarrier of the k-th code. The detected signal vector {right arrow over ({circumflex over (d)}contains the data symbols of all the codes in a concatenated form such that:
where {right arrow over ({circumflex over (d)}(i,k) are the data symbols carried on the k-th code in subset i and have the variable data length of
depending on the spreading factor Qi,k. {right arrow over ({circumflex over (d)}(i,k) can be represented as:
Although the features and elements of the present invention are described in the preferred embodiments in particular combinations, each feature or element can be used alone without the other features and elements of the preferred embodiments or in various combinations with or without other features and elements of the present invention.
This application claims the benefit of U.S. provisional application No. 60/687,834 filed Jun. 6, 2005, which is incorporated by reference as if fully set forth.
Number | Name | Date | Kind |
---|---|---|---|
6219341 | Varanasi | Apr 2001 | B1 |
6961395 | Fuller et al. | Nov 2005 | B2 |
7133646 | Miao | Nov 2006 | B1 |
20020048333 | Ahmed et al. | Apr 2002 | A1 |
20030185179 | Inogai et al. | Oct 2003 | A1 |
20040062317 | Uesugi et al. | Apr 2004 | A1 |
20040120274 | Petre et al. | Jun 2004 | A1 |
20050018598 | Borran et al. | Jan 2005 | A1 |
20060245472 | Pan et al. | Nov 2006 | A1 |
Number | Date | Country | |
---|---|---|---|
20060274842 A1 | Dec 2006 | US |
Number | Date | Country | |
---|---|---|---|
60687834 | Jun 2005 | US |