MXPA98004021A - Determination of signal to noise relationship using digi signal processing - Google Patents

Abstract

The present invention relates to the methods and system for measuring a noise signal of supervisory audio tones (SATs) in radiocommunication systems. The exemplary embodiments involve the extraction of a real component of a SAT received signal, extracting the portion of the SAT signal and checking the remainder to determine a noise component present in the SAT. The noise component can be compared to a measure of the strength of the signal to obtain a proportion of a noise signal.

Description

METHOD AND APPARATUS FOR CALCULATING THE PROPORTION OF THE NOISE SIGNAL IN RADIOCOMMUNICATION SYSTEMS USING AUDIO SUPERVISION TONE MEASUREMENTS Related applications The following patent applications relate to the present application: U.S. Patent Application Serial No. 08 / 561,055, entitled "Supervisory Audio Tone Detection Using Digital Signal Processing" by Omar Ryde et al. And the Application for Patent with Serial No. 08 / 562,085, entitled "Radio Channel Squelching Systems and Methods" by Jafar Rostamy et al., Both applications were filed with the same date as the present application. The information of these related applications are expressly incorporated herein by reference.

Background of the Invention The present invention relates generally directly to radio communication systems, and more particularly, to techniques for determining a noise of a radio channel in a radio communication system, and therefore to a signal to noise ratio (SNR) of this radio channel.

In movable cellular radio systems, it is desirable that a mobile station with a connection established in a radio channel could be able to monitor the quality of the established connection when moving from a cell serving a base station to another cell that serves another base station. The process by which a mobile station maintains an established connection when it moves between cells in a cellular radio system is usually called single-ended. Also very considerably, it would be desirable that a mobile station with a connection established in a radio channel could maintain the connection when it moved within the same cell, even if the radio channel being used were subject to increased interference. If the quality of the The connection established falls below the specified parameters, it would also be desirable to temporarily disconnect it, or automatically disconnect the connection in the case that the loose command or other processing signal does not improve the quality of the connection.

In general, radio-collision is only possible when the radio signals carrying desired information have sufficient signal strength in the receiver and are sufficiently strong in relation to the noise and radio interference signals in the receiver. The noise present in a radio channel and consequently the proportion of the noise signal (S R) depends on particular characteristics of the system, for example, the kind of modulation and receiver that is used. In order to determine whether an established connection should continue on a selected radio channel between a mobile station and a base station, the loose-off and disconnection processes carry out several means on the radio signals on the base attempted and / or in mobile stations.

The first cellular mobile radio systems in public use were analogous systems that were used to transport language or other analogous information. These systems comprised multiple radio channels to transmit the Modulated analog radio signals. In general, the measurements of the signals made during the loose and disconnection processes in these systems were carried out by the base stations. A system of this type is known as the Nordic Mobile Telephone MT 450 system. Another known analog cellular mobile radio system, of particular interest as antecedent to the present invention, is the Advanced Mobile Phone Service (AMPS) mobile radio system. It is used in the United States.

Recently, digital cellular mobile radio systems were designed for public use. Digital cellular mobile radio systems provide digital radio channels for digital transmission or digitized analog information between base stations and mobile stations using digitally modulated radio signals. Digital cellular mobile radio systems can offer substantial advantages, for example, a much higher system capacity per unit bandwidth, in relation to analog cellular mobile radio systems. To achieve these advantages there are certain demands. In particular, channel supervision, loose control and disconnection processes need to be carried out quickly and frequently in relation to conventional analog systems.

In contrast to the introduction of only digital cellular mobile radio systems, such as the GSM system used in parts of Europe, in areas with existing analog cellular systems it has been proposed to introduce digital cellular mobile radio systems that were designed to cooperate with the existing analog cellular mobile radio systems. In this way a large number of customer legacy bases will not suddenly find that their equipment ends up being obsolete. System designers of these hybrid systems believe that the digital portion of the system can be introduced gradually and that over time the amount of digital channels can be gradually increased, while the number of analog channels gradually decreases. In order to provide full compatibility, these dual-mode systems should proceed with both analogue and digital standards that have been adopted, for example, the analog AMPS standards and the IACS.

In addition to an information signal whose force can be easily measured, a radio channel signal comprises a noise component that can also be measured and compared with a corresponding force measurement to provide an indication (SNR) of the signal ratio for he noise. Conventional analogue base stations measure the strength of the channel signal and noise using analog components and periodically provide an indication of the proportion of the signal for channel noise to the network. The quality of the transmission of the channel can be measured based on the indication of the proportion of the signal for the noise.

In AMPS, a supervisory audio tone, abbreviated SAT, is transmitted on analog communication channels. A measure of the strength of the SAT signal is determined to monitor the presence of a channel connection. More specifically, a base station transmits a SAT to a mobile station, which receives the SAT and responds back to the tone of the base station to close the circuit. The reason for transmitting the SAT in AMS is that in a mobile radio-cumulation network of limited interference, there should be some mechanism for the receiving entity (eg, a base station) to identify the transmission entity (eg, a mobile station). ) or at least with an exchange of high probability of exclusion of the transmitting entities without the need for a continuous transmission to a transmission identity. In this way, the base station expects to receive the same SAT that was sent out, that is, on the same frequency. If it is received In the base station a different SAT, then the connection is perceived to be interfered with and can be disconnected. In order to qualify as a valid SAT, the base station must receive it at some predetermined signal strength. In this way the conventional analogue base stations measured the SAT outside using components of analogous sets and provided a report of the strength of the SAT signal to the network periodically, for example the AMPS standard specifies that the strength of the SAT signal is reported when less every 250 ms.

Although dual-mode base stations and mobile stations continue to support analog system functions, such as measuring the strength of the SAT signal strength and SNR determination, the way in which these functions are supported is continually improved to reduce costs and improve quality. With the increased strength of digital signal processors (DSP) system designers are interested in implementing many signal processing techniques previously implemented using analog set components such as DSP routines. The implementation of the DSP has the additional attraction of reducing the number of components, and therefore the size, of the base stations and mobile stations.

Of course the processing of the digital signal also has its limitations. A design transaction that system designers confront when trying to implement analog signal processing techniques such as DSP routines, is the speed of the execution of a routine versus the number of DSP sources, for example, the available amount of millions of instructions per second (MIPS) that are distributed to execute the given routine. Because the processing of the digital signal is not yet cheap enough in relation to the amount of MIPS used in each routine is negligible, the designers of the system were granted the development of innovative digital signal processing techniques that reduce the number of MIPs that are used, so that the digital signal processor can handle as many tasks as possible.

The use of DSP routines to execute tasks previously carried out by the analogous components is not a right proposition. Limited DSO sources given for the operation of a variety of tasks, designers must have the ability to handle the voucher of the converted signals, the processing of these signals, the extraction and / or calculation of the information desired, and the output processing of the desired information and / or signals.

However, as previously alluded to, DSP sources for the performance of these tasks are typically limited in terms of cost, skill and speed. Therefore, digital signal processors should be used in an efficient manner to achieve the advantages of their use, while maintaining the sampling of the required signal, processing and provision of the desired information, as a measure of the proportion of the signal for noise.

Compendium These and other drawbacks and limitations of conventional methods and techniques for measuring, for example, noise and SNR in a radio communication system are overcome in accordance with the present invention. Exemplary modalities describe how these measures can be carried out using digital signal processing techniques, while minimizing the amount of computing power, for example, the MIPs that are used to accomplish these tasks. According to the invention, a quadrature component (I or Q) is extracted from a current of signal samples and is examined to determine the noise present in the signal. A known information component is extracted from the signal samples by leaving a remnant, from which a measurement of the noise signal can be derived.

In an exemplary embodiment of systems and methods embodying the invention, the SNR measurement is combined with the SAT strength measurement to reduce the entire amount of computations that are used to provide these measurements. For example, samples of the SAT signal are taken from where the real component is extracted (in phase or "I"). The extracted real component is filtered at low pass to remove the SAT signal. The variant of the filtered output is used to obtain a measurement of noise. The noise measurement can be compared to a measure of the strength of the SAT signal to provide an SNR indication: Brief Description of the Drawings The foregoing characteristics and advantages and other objects of the present invention can be more easily understood by reading the following detailed description together with the drawings, in which: Figure 1 is a block diagram that generally illustrates the measurement of SAT and that reports in a rediocommunication system in accordance with an exemplary embodiment of the present invention; Figure 2 is a block diagram illustrating a derivation of an imaginary component to provide an indication of the signal noise SAT, y; Figure 3 is a flow diagram of an exemplary embodiment of the present invention.

Details Description As previously described, supervisory audio tone (SAT) is used to monitor connections between base stations and mobile stations for analog radio traffic channels. To generally describe how the SAT is used, Figure 1 illustrates exemplary functional units in a readiocomunication system 100 in the form of a general block diagram. The radio base station (RBS) 106 transmits its SAT. A mobile station (MS) 112 repeats a SAT on the same frequency as was detected in the transmission from the RBS back to the RBS 106 during the time that was connected to the RBS 106 via an analog traffic channel. The strength of the SAT signal and the noise was detected and measured, in accordance with an exemplary embodiment of the invention, by a digital processor (DP) or a digital signal processor (DSP) 110 at the radio base station 106. The The digital signal processor (DSP) 110 reports the strength of the received signal and the noise of the SAT to another processor 108 (sometimes referred to as a regional processor (R?)) in the RBS 106. The processor 108 interprets the strength of the the reported SAT signal and noise, for example, a proportion of a noise signal (SNR) and can compare the SNR with a predetermined threshold. This interpretation is then followed to the network via a central processor (CP) 104 of the mobile switching center (MSC) 102. Because many of the details of the components of the station per se are not related to a discussion of the present invention, the Applicant omitted these details to avoid obscuring the invention. Readers interested in additional details of the base stations, including digital signal processors, are generally directed to U.S. Patent No. 5,295,178, the information of which is incorporated herein by reference.

The systems operating in accordance with the present invention, operate with the premise of extracting the anticipated signal from a received signal, filtering the anticipated portion and examining any remaining signal to determine a quantity of noise present as an indication of noise in the channel of transmission. An exemplary embodiment of the invention uses SAT. The SAT is continuously broadcast along with a voice signal at a known frequency and can be easily extracted. The noise is usually ubiquitous, so the investigation of any bandwidth of a SAT sample from which the SAT was filtered can provide a reliable indication of the amount of noise present in a channel.

Conventionally, analogous components have also been used to measure the out of the SAT signal in the radio base stations. The technique of measuring the strength of the SAT signal applied in these conventional systems can be recreated in a digital signal processing routine by performing a digital or discrete Fourier transformation (DFT) on the received signal at the expected frequency of SAT . The DFT transformation of the received signal is illustrated in Figure 2 in block 202. Those skilled in the art will become familiar with the way in which a digital fourier transformation. In Figure 2, the SAT force is derived in the upper portion £ 02 of the circuit diagram to produce a force average S2. A more complete description of a method and apparatus for determining the aforementioned SAT strength, including the functionality of the unnumbered blocks in Figure 2, is disclosed in the co-pending US Patent Application. previously incorporated Series No. (BDSM 027555-481) entitled "Supervisory Audio Tone Detection Using Digital Signal Processing". However, the particular method and the means for carrying out force measurement is not critical in the present invention, which mainly involves the determination of a noise signal. However, it is of great significance that the repeated step of multiplying the samples of the input signals to extract the real component is carried out just once for both the detection and the SAT measurement, thereby saving a large amount of calculations.

As will be appreciated by one skilled in the art, a measure of signal strength, by any means, can be combined with a noise measurement as provided by systems operating in accordance with the present invention, to produce, for example , a measure of SBR. As long as a signal and noise measurement is made Consistently and compare, the corresponding SBR will provide information indicative of channel quality. With this, the measurement of the strength of the signal can be carried out by any suitable method or apparatus.

Referring again to Figure 2, the current of the input sample is separated into its real components (in phase), I, and into imaginary components (quadrature) Q, in the mixers 204 and 206 respectively. The frequency ? which was chosen for the mixture in blocks 204 and 206 is what was expected back from SAT by the mobile station. A number of different SAT frequencies can be used in a radio communication system to differentiate between neighborhood cell transmissions. For example, three frequencies SAT, 5970, 6000 and 6030 Hz, are used in AMPS.

In accordance with an exemplary embodiment of the invention, the actual component (in phase or "I") 207 extracted from the input and output signal from the mixer 206 is provided for the noise determination block 120 to calculate a noise component. of the SAT input signal. A low pass filter 208 within the determination block of noise 210, filters the actual component 207 to remove the portion of the SAT signal. In accordance with the preferred embodiments, the low-pass filter 208 samples the output of the mixer 207 at a rate, for example, of 1000 Hz, so that the noise to be measured is at a much lower frequency that of the SAT signals and allows the signals to have a frequency that is not greater than 100 Hz to pass, thereby blocking any of the signals 5970, 6000 or 6030. Those skilled in the art will appreciate that the signal is it could sample at any desired frequency, which is consistent with the restrictions imposed by sampling theory. The output 209 of the low pass filter 208 is then used to derive the noise measurement. According to the preferred embodiment of the invention, the noise measurement is carried out by calculating a variant of the leakage of the filter 209. The variant can be calculated in accordance with equation I.

? M2 - (? M) - N2 Equation 1 where M is a signal sample.

Calculating as shown, the variant provides a noise measurement having a value of N-, which can be combined with the force output S2 from block 202, such as Sz / N2, to provide a measure of noise signal proportion . The US Application above-mentioned of Series No. describes a major detail of the evaluation of the strength of the signal SAT, S2, in association with block 202.

The exemplary embodiment of Figure 2 uses a distribution of arithmetic logic devices within block 2122 to perform the calculation of the variant. Within block 212, a square device 211 blocks each sample of the filter output 209. The output of the square values by the square device 211 is accumulated within the device to sum 213. The devices to square and add, 211 and 213 form the first element of Equation 1. Each example of the filter output 209 is also provided and accumulated within the device to add 215. The cumulative value within the device to sum 215 is squared by the pair-wise device 217 to provide a value corresponding to the second element of Equation 1. The output of the device to square 217 is subtracted from the output of the device to add 213 in a 219 adder to provide the measure of noise:. Of course, variant calculation can be carried out in any number of ways, including processing steps in a DSP routine. An exemplary routine of pseudo code is provided below. for (i = o; i <SAT_FRAME_LEN; i ++). { y / n) = aO.x (n) + al. (n-1) + a2x (n-2) + b0. and (n-1) + bly (n-2); y (n-2) = y (n-l); / + a0, al, a2, bO, bl are the coefficients * / of the LP filter y (n-l) = y (n); y (n-2) 0 x (n-l); x (n-l) = x (n); (sampling below) / * Sampling below lKHz . { and (n) 0 = y (n) * 4; noise-add-sar = noise-sum-sgr + force (y (n), 2); noise-sum-sum = noise-sum-sum + y (n) * 4; } yes (SAT_DFT_LEN) / ^ time to compute 2 * I sum-noise_ = noise_sum_sar = force (noise sum sum 2) SAT_DFT_LEN / * SNR = 10 Log.10 (suma_sat / 'sum_ruido) * / A method of conformance to an exemplary embodiment of the present invention is illustrated in FIG. 3. A signal is received by a base station in block 302. A SAT signal is extracted from the receiver signal in block 304. The extracted SAT signal it is mixed with a signal (ie, cost) to extract an actual component of the SAT signal (in block 306). The actual component is then filtered in block 308 to discard the components of the SAT frequency. The variant of the filter output is calculated in block 310, whose variant is an indication of channel noise. The variant can be compared to a measure of strength in the SAT signal in block 312, to determine a proportion of signal noise of the SAT signal.

The SAT signal is continuously present during the transmission of the voice. With what is a very convenient signal to be used for purposes of determining a proportion of the noise signal of a voice channel. This is especially true in view of the possible ability of the measurement of the strength of the SAT signal carried out within the block 202 in accordance with the US application. Series No. cited above and incorporated by reference. It will be appreciated that any other known signal type can be extracted from the received signal and then have its portion of the information, such as the specific SAT, which is extracted in order to extract a difference or remnant that can be used to determine a noise component. . For example, any known or expected signal above the band of the frequency associated with the transmitted language can be extracted from a received signal. The anticipated value can be subtracted from the extracted signal to produce a difference value whose variant is indicative of the noise in the signal.

The exemplary embodiments described above are intended to be illustrative in all respects, and not restrictives, of the present invention. Thus, the present invention is capable of many variations in the detailed implementation that can be derived from the description contained herein by a person skilled in the art. All variations and modifications are considered to be within the scope and spirit of the present invention, as defined by the following claims.

Claims (18)

twenty-one
1. CLAIMS i. A method for measuring a noise component of a received signal using a digital signal processor, the method comprises the steps of: performing a digital fourier transformation on the received signal that includes the components of the real and imaginary extraction of a supervisory audio tone signal; filter the real components to remove the monitoring audio tone signal and provide a filtered output; calculating a sum of a square of the filtered output to provide a first value; calculate a square of a sum of the filtered output to provide a second value, y; subtracting the second value of the first value to provide a variant, whose variant is indicative of the noise component of the supervising audio tone signal.
2. 2. The method of claim 1, further comprising the step of: comparing the noise component with a signal strength to determine a signal of the noise ratio.
3. 3. The method of claim 1, wherein the step of filtering the actual components comprises the step of: Pass the real components through a low pass filter.
4. 4. The method of claim 3, wherein the low pass filter also samples the actual components.
5. 5. A method for measuring a noise component of a known signal comprising the steps of: use a digital fourier transformation computation to extract the real components of the known signal; discard a known portion of the real component with which a remnant is obtained, and; determine a variant of the remnant as a measure of the noise component.
6. 6. The method of claim 5, wherein: the known signal is a supervisory audio tone.
7. 7. The method of claim 5, wherein the step of discarding a known portion further comprises the step of: filtering the supervisor audio tone.
8. 8. A system for measuring the noise component of the supervisory audio tone comprising: means for receiving a sample stream means for extracting an actual component of the monitoring audio tone from the sample stream; means for extracting the monitoring audio tone from the real component thereby providing a remnant; means for determining a variant of the remnant, whose variant is indicative of the noise component in the supervisory audio tone.
9. 9. SI system for measuring the signal of the monitoring audio tone of claim 8, wherein the extraction of the real component is part of a digital fourier transformer.
10. 10 The system for measuring the supervising audio tone signal of claim 8, wherein the means for extracting the monitoring audio tone is a filter.
11. 11. The system for measuring the supervising audio tone signal of claim 8, wherein the variant determining the medium provides a variant by subtracting a square from a sum of the remainder of a sum of a square of the remainder.
12. 12. A base station comprising: 25 means for receiving a SAT signal, and; a digital signal processor (DSP) for measuring a component of the SAT signal noise, where the DSP operates on a real component of the SAT to provide a measure of the signal component of the SAT signal.
13. 13. The base station of claim 12, wherein the digital signal processor performs a digital fourier transformation on the received SAT signal.
14. 14. A method to determine a component of the noise of a signal, the method comprises the steps of: show the signal, extract a real component of the sample; filter the real component to extract a low frequency portion, and; calculate a variant of the low frequency portion to determine a component of the noise of the signal.
15. 15. A method for determining a noise component of a signal, the method comprises the steps of; sample the signal; extracting a real component of each signal sample; filter the extracted component to discard the known portion of information cor. what a remnant is obtained, and; calculate a variant of the remnant to determine a component of the noise.
16. 16. A base station comprising: an input node for receiving a sample of a received signal; a mixer for mixing the sample current of the received signal with a sinusoidal signal at a frequency SAI to extract the components of the actual sample; means to detect the noise in the SAT using the components of the real sample.
17. 17. the base station of claim 15, wherein the SAT frequency is one of 5970, 6000 and 6030 Hz.
18. 18. The base station of claim 15, wherein the means for detecting a signal strength and the means for detecting a noise is a digital signal processor (DSP) implementing software routines that each uses the 27 components of the actual sample drawn to reduce a number of MIPs that are needed to implement SAT detection and measure noise in the DSP.