« PreviousContinue »
SYSTEM AND METHOD FOR MINIMIZING
FREQUENCY OFFSETS BETWEEN DIGITAL
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to digital communication between two stations; and more particularly to a system and method for frequency acquisition in digital 10 communication systems.
Although the present invention is suitable for use in many types of digital communication systems, it is particularly advantageous in cellular communication systems for determining the frequency offset between a 15 cellsite and the local oscillator of a mobile receiver; and is described in connection therewith.
2. Discussion of Related Art
A cellular communication system is a mobile telephone service wherein radio coverage is divided into 20 cells, each of which may cover an area in the neighborhood of one to two square miles. Each cell is assigned a number of available radio frequencies. The same frequencies or channels used in one area or cell are also used for areas that are spatially separated from one 25 another. A mobile telephone station transmits and receives control and voice communication information to and from a base station, commonly referred to as a cellsite, located within the same cell. The base stations are controlled by a cellular system switching and con- 30 trol network that provides connection with the worldwide telecommunications network.
A call in progress is not interrupted as the mobile station travels from one cell location to another, since the system provides for automatic reassignment to an 35 available time slot of an available channel within the other cell commonly referred to as a handoff.
In order to provide superior non-interfering communication and compatibility among many different base stations and mobile stations in different parts of the 40 world, various operational and material specifications and standards were developed, which all suppliers and users are obliged to follow. For example, a mobile station that operates in the digital mode is required to use an RF band which is divided into two separate twenty- 45 five Mhz wide segments, each consisting of eight hundred thirty-two channels. The first segment contains the mobile station transmit channels, and the second segment contains the mobile station receive channels. Thus, each transmit and receive channel is approxi- 50 mately thirty kilohertz or kilocycles in width.
Each channel has a frame format; that is, each channel radiates a succession of frames, each of which has a duration of forty milliseconds, and constitutes one cycle of a regularly recurring series. Each frame has six time 55 slots, and each slot has one hundred sixty-two data symbols and a duration of 6.67 milliseconds, for exampie.
The term frequency offset as used herein is the difference between the frequency of the local oscillator in the 60 mobile receiver and the frequency transmitted by the base station of the cell. In order for a demodulator in a digital cellular phone to successfully acquire a signal from a base station, the frequency offset should be less than eight hundred Hz, in that synchronization of the 65 frame requires that the frequency offset be within eight hundred Hz in order for the bit timing to be correct. Any greater offset will result in the call being dropped
if there is a transfer to another cell. This frequency offset should be reduced to within plus or minus two hundred Hz(minus the transmit/receive difference of forty-five Mhz). This two hundred Hz offset is also the maximum initial frequency offset which the demodulator automatic frequency control (AFC)loop is able to tolerate.
A frequency offset between a base station and a mobile receiver can occur for several reasons, such as difference in ambient temperature, aging of the components over a period of time, and assignment of a channel by a new base station, for example. Also, in digital to digital communication where two cellsites are at the limit of their tolerance, a handoff from one to the other can appear to be offset by as much as 610 Hz. During analog to digital handoffs, the initial frequency offset may be much larger. For example, the initial frequency in the worst case may be offset by 4500 Hz because of the tolerance of the voltage controlled oscillator.
Therefore, in order to meet the required maximum frequency offset of ±200 Hz, it is necessary that the frequency of the mobile station can be varied over a certain range to minimize any frequency offset. The carrier frequency of the receiver is determined by a voltage controlled local oscillator (VCXO) that is tunable over a certain range, such as forty-five hundred Hz. Prior to the present invention, various systems were provided for controlling the voltage of a VCXO to vary the frequency of the radio receiver. Typically, samples of the received waveform were obtained and correlated, with the VCXO being tuned in accordance with the results. If, after tuning, the frequency offset was still excessive, more samples would be taken, and the control voltage of the VCXO again would be varied by a certain amount. This process would be repeated until the offset frequency came within the desired limits.
In U.S. Pat. No. 4,644,561, a frequency acquisition routine is described that takes advantage of the period of time during which there is no transmission from the base station. At the expiration of this time the base station transmits an unmodulated carrier signal, which causes the IF mixer of the receiver to output another sine waveform whose frequency is proportional to the difference between the VCXO and the base station frequency. The modem program samples the I,Q channels at certain intervals and determines the phase change for each interval, puts it through a low pass filter and sends it as a correction word to control the VCXO. Frequency acquisition is achieved when the phase change becomes lower than a certain level.
Although suitable for the purposes intended, the methods and systems for frequency acquisition prior to the present invention tended to require costly precision voltage controlled oscillators, and the receipt of multiple frames of data in order to effect proper frequency acquisition.
SUMMARY OF THE INVENTION
One of the objects of the present invention is to provide a system and method of frequency acquisition that quickly determines the frequency offset between a base station and its own local oscillator when the mobile station is first introduced into a new cell.
Another object of the present invention is to provide a system and method of frequency acquisition that is capable of minimizing the effect of multipath conditions and fading during a handoff.
Still another object of the present invention is to provide a system and method of frequency acquisition which is not configured to utilize a plurality of frames of information in minimizing the offset frequencies.
A further object of the present invention is to provide 5 a system and method of frequency acquisition that is able to utilize a more economical voltage controlled local oscillator that permits greater tolerance between the center frequency of the carrier and the frequency of the local oscillator of the receiver. 10
Additional objects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objects and advantages of the invention will be realized and attained 15 by means of the elements and combinations particularly pointed out in the appended claims.
To achieve the objects and in accordance with the purpose of the invention, as embodied and broadly described herein, the system of the invention for mini- 20 mizing frequency offsets between digital communication channels, comprises a matched filter responsive to in-phase(I) and quadrature phase(Q) components of each of a plurality of data symbols for generating a corresponding predetermined series of pairs of I,Q sym- 25 bols; a complex correlator responsive to the I,Q symbols for determining iteratively the correlation of the series of pairs of I,Q symbols with a predetermined pattern of symbols and outputting a maximum frequency value for each iteration; a pattern rotator for 30 changing an apparent frequency of the predetermined pattern of symbols for each iteration of the correlation determination; a peak detector responsive to a predetermined number of iterations of correlation determination for detecting a peak output from the number of maxi- 35 mum frequency values; a parabolic interpolator governed by the detection of the peak value for generating a signal having a value corresponding to an estimate of the frequency offset; and a voltage controlled oscillator responsive to the generated signal for varying the fre- 40 quency of a receiver.
In another aspect, the method of the invention for minimizing a frequency offset between a base station transmitter and a mobile receiver of a cellular communication system where each mobile station includes stored 45 data symbols corresponding to a known synchronization pattern for receiving a communication, where each frequency channel of communication includes a series of frames, each frame includes a series of time slots, and each time slot includes a synchronization pattern of a 50 predetermined series of data symbols, and each symbol of the pattern occurs at a symbol time interval, comprises storing in the receiver, I,Q components of a predetermined total number of data symbols of a frame transmitted by the base station; correlating iteratively a 55 predetermined number of times the total number of stored data symbols with the known synchronization pattern of the receiver; modifying the stored synchronization pattern of the receiver to correspond to a pattern having an apparent frequency offset from a previous 60 apparent frequency for each correlation determination iteration; detecting a maximum value for the data symbols of each iteration; calculating a value corresponding to an estimate of the frequency offset in accordance with selected ones of the detected maximum values of 65 the predetermined number of iterations, and changing the frequency of the receiver in accordance with the calculated value.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only, and are not restrictive of the invention as claimed.
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate one embodiment of the invention and together with the description, serve to explain the principles of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic block diagram of a frequency acquisition system in accordance with one embodiment of the invention;
FIG. 2 is a diagram representative of the output of the complex correlator of FIG. 1 with minimum correlation between the known synchronization pattern of the receiver and the frame of data transmitted by the base station;
FIG. 3 is a diagram representative of the output of the complex correlator of FIG. 1 with some correlation between the known synchronization pattern of the receiver and the frame of data transmitted by the base station;
FIG. 4 is a diagrammatic representation of an array of maximum correlation values corresponding to successive iterations;
FIG. 5 is a parabolic curve with correlated values for distinct frequency offsets of the receiver's synchronization pattern;
FIG. 6 is a flowchart illustrating the steps for the complex correlation of the data symbols in accordance with the present invention;
FIG. 7 is a flowchart illustrating the steps for the parabolic interpolation of the maximum values of each iteration of correlation determination;
FIG. 8 is a schematic representation of a frame of a communication channel for a cellular communication system; and
FIG. 9 is a schematic representation of a slot of the frame of FIG. 8.
DESCRIPTION OF THE PREFERRED
Prior to discussing the system and method of the present invention; a description of the frame and slot protocol utilized in connection with a cellular system will be briefly described in connection with FIGS. 8 and 9 in order to better understand the invention. FIG. 8 illustrates a frame of information generally referred to at 10, which is transmitted every forty milliseconds or at a rate of 25 frames per second. Frame 10 has six slots, referred to as slots 12. Of the slots 12, two slots A may be used by one mobile station, two slots B another, and two slots C by a third station or subscriber for carrying on conversations simultaneously. Each slot represents an individual burst of RF energy of a duration of 6.67 milliseconds. Referring to FIG. 9, one of the slots 12 has several fields, a synchronization field SYNC, two data fields of one hundred forty-two bits each, and a coder digital verification color code CDVCC. The synchronization pattern or SYNC portion permits the mobile unit to acquire the cell data master timing reference. The present invention utilizes that first 20 millisecond or 486 symbols of each frame that includes 3 slots each having a known sync portion.
Reference will now be made in detail to the present preferred embodiment of the invention, an example of
which is illustrated in the accompanying drawing. The system of the present invention is preferably implemented in software on a single fixed point digital signal processor of the well-known type used for demodulation of a digital data stream.
Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.
The system of the present invention for minimizing frequency offsets between digital communication channels includes a receiving modem for a mobile communication system which comprises a down converter, an analog to digital converter, a matched filter, an automatic gain control, a complex correlator, a sychronization pattern, a pattern rotator, a peak detector, a para- ■ bolic interpolator, and a voltage control oscillator.
As herein embodied, and referring to FIG. 1, the system of the present invention is generally referred to at 14, and comprises an antenna 15 for receiving an RF frequency from a base station or cellsite (not shown) which is converted to an IF frequency at down converter 16. The IF frequency is output to an analog to digital converter 18 which outputs a pair of in-phase (I) and quadrature phase (Q) symbol components on lines 20 and 22. I and Q data samples are input to a digital matched filter 24 at the rate of two complex samples for each symbol time. Matched filter 24 includes a finite impulse response filter which samples the I,Q components at twice the sampling rate of the analog to digital converter 18. Thus, the output of the matched filter is two I,Q pairs of data symbols for each symbol time of the system. An automatic gain control circuit 26 computes the gain by a base band demodulator in order to bring the IF signal level into the converter 18 to a nominal level. The automatic gain control is computed by looking at the magnitude of the phase vector after digital filtering by filter 24. The automatic gain control value from block 26 is time averaged and output to down converter 16 in order to filter the response and reduce any noise effect.
The system of the present invention comprises a complex correlator responsive to the I,Q symbols for determining iteratively the correlation of the series of pairs of I,Q symbols with a predetermined pattern of symbols. As the correlator of the known synchronization pattern is a function of two variables time t, and frequency f C(t,f) is a three-dimension parabola with its maximum set at C(0,0). If one variable is fixed, and the second is varied, C(t,f) is a parabola in two dimensions. The frequency acquisition algorithm of the present invention utilizes the fact that for a fixed timing offset, the correlation peak will be proportional to the amount of the frequency offset between the transmitter and receiver. If the peak at various induced frequency offsets is searched, the correlator will produce a peak at the off- 55 set which is closest to the transmitter frequency.
As herein embodied, complex correlator 30 is used to determine the correlation of the incoming data stream . out of matched filter 24 with a known synchronization pattern or derivative thereof represented by block 36. Correlator 30 is run at twice the symbol rate in order to correlate the I,Q, values out of the matched filter 24. Therefore, correlator input sequences are formed Wn, Qeven, lodd, and Qodd- Correlations are done on the even and odd sequences separately. Correlator output energies, ... and Eodd are both measured every symbol time. The energies are used to normalize the correlation results, since the correlator outputs are amplitude de
pendent. The detection parameters FCVcn and Fodd indicate the strength of the correlation at each half symbol.
Even and odd correlator products are computed each symbol and half symbol time, and output to peak detector 44. Assuming that the synchronization pattern stored in box 36 has a length of 15 symbols, the following correlator products are created.
In the products referenced to above the index k specifies the I,Q samples that correspond to the symbol time, thus, k is going to be running between 1 and 486 times. Thus, there is obtained a sampling of 486 pairs of I and Qevens and 486 pairs of I and Qodds- a-i and /3; are the rectangular coordinates of the stored synchronized pattern known to the mobile station. The i runs from 1 to 15 and corresponds to the symbols of the known synchronization pattern. Thus, every half symbol time a correlation product will be computed which is referred to as either odd or even I,Q components. C(I) of the in-phase products refers to correlator output for even symbols and odd symbols. Similarly, C(Q) refers to a correlator product for either the phase quadrature odd or even components, respectively. Since the correlator is complex and the inputs are complex, the correlator output will be a complex number. Therefore, when the correlator is run there will be four different outputs; that is, I and Q components for both the even and odd samples. The energy estimates are calculated to normalize the numbers to eliminate false results. For example, if there were a high input signal, a high output number would be obtained with little correlation, also, with a very low input signal, a very low output of the correlator could occur even though both of them were created by the synchronization pattern. Thus, the energy calculation permits the outputs to be directly proportional to the amplitude of the I,Q symbols coming in over inputs 32 and 34; thus, the normalization permits the process to ignore the amplitude of the I,Q symbols and act upon the phase of the symbols. This normalization provides additional protection that is provided by automatic gain control circuit 26 for minimizing the effect of a fading environment. The energy estimates are computed as follows:
The detection parameters Feve„ and Fodd at the output of the complex correlator are computed as follows: