METHOD AND SYSTEM FOR DETECTING SIGNALS
TECHNICAL FIELD OF THE INVENTION
This invention relates generally to the field of communication systems and more specifically to a method and system for detecting signals.
BACKGROUND OF THE INVENTION
The rising use of wireless communications systems has led to the demand for increasingly effective and efficient signal detection. In one-way communications, a transmitter network transmits signals to a user device, for example, a paging device. In two-way communications, the user device transmits signals back to the transmitter network. User devices often have low transmission power that cannot send strong signals across long distances having obstacles and non-relevant signals in the propagation path. Moreover, signals from other wireless networks may interfere with paging transmissions. In some locations, for example, the spectral separation between cellular telephone transmissions and paging transmissions is not large enough to prevent interference.
One technique to improve signal detection uses two or more receiver antennas. The antennas are arranged to receive individual decorrelated signals that, when properly combined, yield the intended transmission. Prior systems then combine signals from multiple antennas to improve signal detection. While these approaches have provided improvements over prior approaches, the challenges in the field of communication systems have continued to increase with demands for more and better techniques having greater
effectiveness. Therefore, a need has arisen for a new method and system for detecting signals.
SUMMARY OF THE INVENTION In accordance with the present invention, a method and system for detecting signals are provided that substantially eliminate or reduce the disadvantages and problems associated with previously developed systems and methods.
According to one embodiment, a system for detecting signals is disclosed that includes two antennas that receive signals. A processor calculates a cancellation solution and an enhancement solution from the signals. The processor uses the cancellation or the enhancement solution to detect the signals. In particular embodiments, the processor generates the cancellation and enhancement solutions using an eigenvector of the covariance of the signals. The processor may also generate a hybrid solution to detect the signals .
According to another embodiment, a method for detecting signals is disclosed. A signal is received, and a cancellation solution and an enhancement solution are generated from the signal samples. One of the solutions is used to detect the signals. In particular embodiments, an eigenvector of the covariance of the signals is used to compute the cancellation and the enhancement solutions. Also, a hybrid solution is generated and used to detect the signals .
Technical advantages of the present invention include more accurate signal detection. Existing systems that use either signal enhancement or interference cancellation do not provide the optimal antenna combination. This communication system detects signals selectively using
signal enhancement, interference cancellation, or a combination of the two in order to find the optimal solution. The system also incorporates a method for searching for increasingly better solutions to find the optimal combination of antennas. By determining the optimal combination, the method can detect weaker signals travelling over long distances with obstacles and interference along the propagation path. Thus, the present invention allows for effective two-way transmission of signals in wireless communication systems. The system may detect signals using an efficient dominant eigenvector calculation, thus increasing computing efficiency and allowing for the evaluation of individual signal packets.
Other technical advantages are readily apparent to one skilled in the art from the following figures, descriptions, and claims.
BRIEF DESCRIPTION OF THE DRAWINGS
For a more complete understanding of the present invention and for further features and advantages, reference is now made to the following description, taken in conjunction with the accompanying drawings, in which:
FIGURE 1 is a block diagram of one embodiment of a system for detecting signals in accordance with the present invention;
FIGURE 2 is a block diagram of a processor that may be used in the system;
FIGURES 3A-3D illustrate a set of solutions generated by the system for detecting the signal; and FIGURE 4 is a flowchart demonstrating one embodiment of a method for detecting signals in the system.
DETAILED DESCRIPTION OF THE DRAWINGS
FIGURE 1 is an embodiment of a system 100 for detecting signals. Data enters system 100 via, for example, e-mail, telephone, modem, or private data networks. The data is sent through a communication network 103 (such as satellite, microwave, or other suitable wireline or wireless network, or a combination of the preceding) controlled and maintained by network controller 101 to a transmitter 119. Transmitter 119, in turn, transmits the information to a user device 112. User device 112 transmits data packets back to a receiver 102. System 100 contemplates transmitter 119 with receiver functionality or receiver 102 with transmitter functionality.
The data packets received without error are assembled at receiver 102 and sent to an appropriate recipient coupled to network 103. If any data packets are received with errors on either the forward or the reverse path in a two- way system, the packets may be retransmitted until no errors are present. In the particular embodiment of full two-way messaging, the user has the additional benefit of being able to enter custom messages on a portable user device to initiate or answer pages, e-mail, faxes, or telephone messages .
Multiple antennas 114 and 116 may be used to detect signals. Signals sx 104 and s2 106 received by multiple antennas 114 and 116 at receiver 102 are properly combined to receive the intended transmission. One method combines signals s2 104 and s2 106 received from multiple antennas 114 and 116 in order to enhance the signal. The signal enhancement method works well when the noise is very small compared to the signal. Another method combines the signals s2 104 and s2 106 from multiple antennas 114 and 116 in order
to cancel the interference. The interference cancellation method works well when the interference is much stronger than the signal.
Sometimes, the signal enhancement method enhances unwanted interference. Also, the interference cancellation method may reduce the desired signal. Moreover, in many situations, pure interference cancellation or pure signal enhancement does not provide the optimal antenna combination.. A signal combination that works well for a given message or data packet will not necessarily work well for other packets. For example, pager messages are short and each has a potentially different signal-to-noise ratio, interference, and direction of origination.
To improve signal detection, receiver 102 combines spatially diverse or otherwise decorrelated signals using a cancellation solution and an enhancement solution. Receiver 102 selects the solution that yields an optimal combination in order to improve performance. Specifically, user device 112, which may be, for example, a pager, telephone, personal digital assistant (PDA) , or other user device, transmits a signal which is received by at least two antennas at receiver 102. Receiver 102 may be, for example, a paging receiver, cellular base station, or other wireless component that receives signals from user device 112. Receiver 102 may also be affected by noise from other transmitters 119, and by thermal noise produced by electronic circuitry within receiver 102.
In a particular embodiment, receiver 102 includes a first antenna 114 and a second antenna 116. Receiver 102 may also include more than two antennas to provide further decorrelated signals for processing. Antenna 114 receives signal sx 104 and noise vx 118, and antenna 116 receives
signal s2 106 and noise v2 120. Signals s2 104 and s2 106 represent at least a portion of a packet transmitted by user device 112. A packet includes any portion of information transmitted by user device 112, including the payload, preamble, or separate interference periods. Receiver 102 includes a modem 122 coupled to antennas 114 and 116. Modem 122 has a variety of components or stages to perform radio frequency (RF) and intermediate frequency (IF) filtering, downcoverting, and other processing. In the non-limiting illustrated embodiment, modem 122 includes a filter 124, an automatic gain control (AGC) 126, a down-converter 128, and a sampler 130.
In this embodiment, modem 122 receives signals 104, 106, 118, and 120. Filter 124 filters the received signals 104, 106, 118, and 120, AGC 126 adjusts the gain of the signals, and down-converter 128 converts the frequency of the signals to a lower frequency. Sampler 130 samples signal s2 104 and noise v2 118 received by antenna 114 and outputs sample x 132. Sampler 130 also samples signal s2 106 and noise v2 120 received by antenna 116 and outputs sample x2 134. A signal evaluator 136 evaluates samples Xj 132 and x2 134 for errors. For example, a checksum may be used to evaluate the preamble, payload, and/or other portion of a signal packet. In a particular embodiment, signal evaluator 136 divides a packet into a number of interference periods and determines any errors on an interference period basis. Interference period analysis of the signal reduces further processing load when receiver 102 receives most of the packet without errors. If samples x 132 and x2 134 are acceptable, signal evaluator 136 sends the samples to an application 140 for further processing. Application 140 extracts the data, and
provides higher level signal protocols and processing to present data in an appropriate format for communication network 103, network controller 101, the intended recipient, or other components in system 100. If samples λ 132 and x2 134 are not acceptable, signal evaluator 136 passes the signals to a processor 150 for further processing and detection. Processor 150 computes a cancellation solution, an enhancement solution, and, in certain cases, a hybrid solution using samples x2 132 and x2 134. Processor 150 evaluates the solutions and passes an optimal solution zDpt 151 to application 140 for further processing.
FIGURE 2 is an embodiment of processor 150 that may be used in accordance with a system and method for detecting signals. Processor 150 can be one or more digital signal processors, microprocessors, controllers, or other hardware, software, or combination. Processor 150 includes a signal combiner 202, a memory 204, a solution evaluator 206, a hybrid solution generator 208, and a counter 210. The components of processor 150 may be embodied in software, hardware, or a combination of hardware and software at one or more locations. Also, it should be understood that processor 150 may include more or fewer components or modules to accomplish the signal detection features described below.
Memory 204 stores a noise variance σ2 2 of antenna 114 and a noise variance σ2 2 of antenna 116 and communicates the variances to signal combiner 202. Signal combiner 202 calculates a cancellation solution zcan 203 using samples xx 132 and x2 134. Solution evaluator 206 receives cancellation solution zcan 203 from signal combiner 202 and determines whether zcan 203 can be used to detect the signal. Solution
evaluator 206 performs similar functions to assess the signal as signal evaluator 136. If solution zcan 203 is acceptable, solution evaluator 206 communicates solution zcan 203 as optimal solution zopt 151 to application 140. If solution zcan 203 is not acceptable, signal combiner 202 calculates an enhancement solution zenh 205 using samples x2 132 and x2 134 and noise variances σ2 2 and σ2 2. Again, solution evaluator 206 determines whether enhancement solution zenh 205 can be used to detect the signal, and communicates solution zen 205 as optimal solution zopt 151 to application 140 if solution zenh 205 is acceptable. Processor 150 may also perform the generation and evaluation functions in reverse order (enhancement and then cancellation) or in parallel . If neither cancellation solution zcan 203 nor enhancement solution zenh 205 is acceptable, signal combiner 202 communicates solutions zcan 203 and zenh 205 to a hybrid solution generator 208, which generates a hybrid solution Zhyb 209 from solutions zcan 203 and zenh 205. Alternatively, hybrid solution generator 208 may generate zhyb 209 using the original sampled signals x∑ 132 and x2 134. Solution evaluator 206 determines whether hybrid solution zhyh 209 can be used to detect the signal, and communicates the solution Z yb 209 as optimal solution zopt 151 to application 140 if solution 209 is acceptable. If hybrid solution zhyb 209 is not acceptable, processor 150 increments counter 210 and generates additional hybrid solutions until an acceptable solution is generated or until processor 150 performs a specified maximum number of iterations. Hybrid solution generator 209 generates hybrid solutions so that each successive solution moves closer to an optimal solution yielding an optimal combination of antennas, thus improving
signal detection. After computing a solution zopt 151 that can be used to detect signals s2 104 and s2 106, processor 150 communicates the solution zopt 151 to application 140.
FIGURES 3A-3D illustrate a set of solutions generated by system 100 for detecting signals xλ 304 and x2 306. FIGURE 3A is a graph illustrating vector representations of signal x2 304 and signal x2 306 received for example by spatially diverse antennas 114 and 116 of receiver 102. FIGURE 3B is a graph illustrating signal enhancement, where signal x2 306 is transposed and its amplitude optionally adjusted to yield vector x2 ' 308. Vector x2 ' 308 is added to signal x2 304 to yield vector xenh 310. FIGURE 3C is a graph illustrating interference cancellation, where vector x2 ' 308 is subtracted from signal xλ 304 to yield vector xcan 312. FIGURE 3D is a graph illustrating a hybrid solution, where vector xcaπ 312 is transposed and its amplitude optionally adjusted to form vector xcan ' 314. Vector xcan ' 314 is added to vector xenh 310 to yield a vector xhyb 316. Alternatively, processor 150 may generate vector xhyb 316 using original sampled signals x2 304 and x2 306. Processor 150 generates one or more of these solutions using appropriate scaling and adjustments described below to detect signals in system 100.
FIGURE 4 is a flow chart illustrating one embodiment of a method for detecting signals in system 100. In general, the method calculates a cancellation solution, an enhancement solution, and, in certain cases, a hybrid solution in order to detect a signal. Specifically, the method begins at step 402 where antenna 114 receives signal s2 104 and noise v2 118, and antenna 116 receives signal s2 106 and noise v2 120. Modem 122 of receiver 102 processes the received signals at step 404. For example, filter 124 filters the signals, AGC 126 adjusts the gain of the
signals, and down-converter 128 converts the signals to a lower frequency. Sampler 130 samples signals 104, 106, 118, and 120 at step 406. Sampler 130 samples signal s2 104 and noise v2 118 received by antenna 114 to yield sample Xx 132. Sampler 130 also samples signal s2 106 and noise v2 120 received by antenna 116 to yield sample x2 134.
The mathematical model of the received signals for this illustrative two antenna example is given by Equation (1) :
x (j) = a s (j) + v (j) , j = l,2, ... , N (1) where :
(j) ' , or is the antenna array sample vector of the
X CO received signal; j is the sample index;
1 refers to antenna 114;
2 refers to antenna 116; a, is the array response vector to the signal;
s (j ) is a signal sample;
Ns is the number of samples in a packet.
Equation (1) and the following computations use a two antenna, two signal embodiment. However, system 100 and the following computations support any suitable number of antennas and samples to detect at least a portion of a packet transmitted by user device 112.
The elements of xfj , a, s (j ) , and v (j ) are complex numbers representing the quadrature components of the
received signals, the array response vector, the transmitted signals, and the noise vectors, respectively. Moreover, the
vector x (j ) represents the received signals after filtering, automatic gain control, frequency down-conversion, sampling, and appropriate decomposition into the quadrature components .
Signal evaluator 136 checks the bits of samples x± 132 and x2 134 to see if the signals are detectable at step 408. Signal evaluator 136 performs this analysis using checksums or other appropriate technique on the preamble, payload, and/or one of a number of interference periods of a received packet. If the samples are acceptable, signal evaluator 136 communicates samples x2 132 and x2 134 to application 140 at step 410 for further processing, and the method terminates. If the samples are not acceptable, the samples are sent to signal combiner 202.
Signal combiner 202 computes the covariance of the samples at step 412. The sample covariance matrix is given by Equation (2) :
x 1 ^ Xl CO Xl CO Xl CO X2 CO (2) s J=l X2 CO Xl CO X2 CO 2 CO
1 w- -, _,H _, _,H _, _,H _, _,H
— ∑ a sCO s* () a + v(j) v (j) + a J) v (j) + v() a s* (j)
where mH represents the conjugate transpose of matrix m, and y* is the conjugate of a complex value y. If the correlation between noise v2 118 and v2 120 and the samples x± 132 and x2 134 is negligible with respect to signal auto-
correlation and noise auto-correlation, then the approximation given by Equation (3) holds:
aιaι aia2 2 a + ∑ = p + ∑π ∑.
( 3 ) a2a1 a2a2 γ Σ 22 where
is the average power of the signal ;
is a measure of noise; and
Signal combiner 202 decomposes the covariance matrix R of the samples determined at step 412 into eigenvectors and eigenvalues at step 414. Matrix R may be decomposed using
Equation (3) . Alternatively, matrix R may be decomposed using singular value decomposition, as shown in Equation (4) :
R=USU
H (4) where
is the strength of the signal; and
Matrix U contains the eigenvectors u2 and u2 of covariance
matrix R. Dominant eigenvector u2 is the signal subspace,
where —^ gives the estimated ratio of the dominant signal u,
in antenna 114 and that of antenna 116. Eigenvector u2 is the noise subspace. Matrix S contains the eigenvalues of
R . The eigenvalues are real numbers, and Sn≥S22- If the matrix Σ, described by Equation (3b), is diagonal, then Equation (5) holds:
where is an unknown complex scalar. That is, the
decomposition of matrix R provides an estimate of vector a up to a scalar. Matrix Σ is diagonal if there are no interfering signals, except for thermal noise, as expressed by Equation (6) :
Matrix Σ is approximately diagonal if interfering signals are negligible and the noise consists primarily of thermal noise. Nevertheless, Equation (5) yields a good
approximation when the noise vectors v (j ) are very small
compared to the signal vectors as(j) . If this is the case,
dominant eigenvector u2 can be used to enhance the signal by combining the samples from antennas 114 and 116. Equation (5) also yields a good approximation when the noise vectors contain interference that is much stronger than the signal .
If this is the case, dominant eigenvector u can be used to cancel out the interference from antennas 114 and 116. Using the dominant eigenvector results in efficient calculation of solutions, which allows for the evaluation of individual signal packets, thus improving signal detection. Signal combiner 202 calculates cancellation solution
Zcan 203 at step 416 from the eigenvector u2 determined at step 414. Cancellation solution zcan 203 is computed using Equation (7) :
u.. Since —^- gives the estimated ratio of the dominant signal
in antenna 114 and that of antenna 116, eigenvector Ui itself can be used to compute a suitably accurate solution. By using only the dominant eigenvector to evaluate a signal, computing efficiency is increased, which allows receiver 102 to detect signals on a per packet basis, or even per interference period basis, as the signals are received.
067056.0103
15
Solution evaluator 206 evaluates whether cancellation solution zcan 203 can be used to detect the signal at step 418. If solution zcan 203 is acceptable, solution evaluator 206 communicates the solution to application 140 at step 410, which uses the solution to detect the signal, and the method terminates.
If solution zcan 203 is not acceptable, signal combiner 202 computes enhancement solution zenh 205 at step 422 from the eigenvectors determined at step 414. Alternatively, enhancement solution zenh 205 may be computed before cancellation solution zcan 203, or enhancement solution zeπh 205 and cancellation solution zcan 203 may be computed in parallel. Enhancement solution zenn 205 is computed from signal samples x2 132 and x2 134 using Equation (8) :
where <τ, and σ2 represent the noise variances of the first antenna 114 and second antenna 116, respectively. Solution evaluator 206 evaluates whether enhancement solution zenh 205 can be used to detect the signal at step 424. If solution Zenh 205 is acceptable, solution evaluator 206 communicates the solution to application 140 at step 410, which uses the solution to detect the signal, and the method terminates. If neither the cancellation or enhancement solution is acceptable, hybrid solution generator 208 initializes hybrid factors to compute a hybrid solution from cancellation solution zcan 203 and enhancement solution zeπh 205 at step 428. The hybrid solution is computed using Equation (9) :
ZhyfcG') = < ZcanU) + P^CO ( 9 >
where scalar values and β are the hybrid factors. Scalars and β may be bound by enhancement solution zenh 205 and cancellation solution zcaπ 203. Alternatively, the hybrid solution may be computed from signal samples Xi 132 and x2 134 using Equation (10) :
zhyb(j) = axM + &x2 ti) (10)
where scalar values α and β are the hybrid factors. Hybrid solution generator 208 generates hybrid solution Z yb 209 at step 430 using either Equation (9) or Equation (10) and suitable hybrid factors. Solution evaluator 206 evaluates whether hybrid solution Zhyb 209 can be used to detect the signal at step 432. If the solution is acceptable, solution evaluator 206 communicates the solution to application 140 at step 410, which uses the solution to detect the signal, and the method terminates.
If hybrid solution zhyb 209 is not acceptable, counter 210 increments an iteration count at step 436, and processor 150 determines whether the signal is to be declared undetectable by determining whether the iteration count has exceeded a specified maximum number of iterations at step 438. If the iteration count exceeds the specified maximum number of iterations, processor 150 discards the signals and reports to application 140 that the signals are undetectable at step 442, and the method terminates. If the iterations count does not exceed the specified maximum number of iterations, hybrid solution generator 208 adjusts the hybrid factors at step 440. Generator 208 adjusts the hybrid factors by, for example, adjusting the phase and the amplitude values of the solution by predetermined increments within a range defined by the phase and the amplitude values
of cancellation solution zcan 203 and enhancement solution zenh 205. The method returns to step 430 and repeats steps 430 to 442 until a hybrid solution zhyb 209 provides an acceptable solution or until a maximum number of iterations have been performed, and the method terminates. The method searches for a combination of antenna outputs until it obtains a combination that results in acceptable signal detection.
Although an embodiment of the invention and its advantages are described in detail, a person skilled in the art could make various alternations, additions, and omissions without departing from the spirit and scope of the present invention as defined by the appended claims.