HK1114718A - Detection of signal modulation format type - Google Patents

Description

Detection of signal modulation format type

Technical Field

The present invention relates generally to mobile phone technology and, more particularly, to detection of signal modulation format types.

Background

The current standard of popularity for digital mobile phone technology is global system for mobile communications (GSM), which is a second generation (2G) mobile phone system. One expansion enhancement to the GSM communication system that is widely accepted is enhanced data rates for global evolution (EDGE). EDGE technology is capable of increasing data throughput to 384kbps and conforms to the international telecommunication union, third generation (3G) network standard. Services related to 3G telephony include the transfer of voice data (telephone calls) and non-voice data (e.g., email, instant messages, etc.).

EDGE technology transmits and receives signals using both Gaussian Minimum Shift Keying (GMSK) and 8-phase shift keying (8PSK) modulation formats. As is well known in the art, GMSK is a digital modulation technique using phase shift keying in which the information signal can have two possible completely different phase offsets. As is well known in the art, 8PSK is also a digital modulation technique using phase shift keying in which the information signal has eight possible completely different phase offsets. The GMSK technique produces one bit word/symbol for each phase change, while 8PSK produces three bit words/symbols for each phase change (phase comparison with the reference waveform).

In cellular mobile telephone systems employing EDGE technology, EDGE-compatible hardware and software are included in both the base station (e.g., transceiver unit) and the mobile terminal (e.g., cellular telephone) for modulating and demodulating signals employing the EDGE offset keying scheme. The signal is typically modulated and transmitted as data blocks (elementary units of transmitted data), one of which consists of four data bursts. All four data bursts of a data block are typically modulated with the same format type (GMSK or 8PSK), so that each data block of the signal has a uniform modulation format type.

Since EDGE-compliant base stations transmit both GMSK and 8PSK signals, in order to properly demodulate the signals, the mobile terminal must be able to detect the modulation format (GMSK or 8PSK) of the received signal. Typically, modulation format type detection is accomplished by examining each burst of a received signal data block individually and determining the modulation type of that burst. After the modulation type of the burst is determined, the burst is treated (i.e., demodulated) as a burst having the determined modulation format, and so on. A next determination is then made separately for the next burst of data blocks and processed accordingly. Therefore, the modulation type of one burst is generally determined independently of the other bursts of the same data block and then processed independently according to the determination result.

However, conventional methods of determining the type of modulation format of the data burst of the received signal can cause a large number of detection errors (i.e., errors in determining the type of modulation format), particularly in the case of low signal strength (i.e., in the case of low received signal-to-noise ratio). Therefore, a more accurate method is needed to detect the received signal modulation format type, especially in cases where the signal strength is low and the probability of detection error is much higher.

Disclosure of Invention

Embodiments disclosed herein address the above stated needs by providing more accurate biasing algorithms and apparatus for detecting the modulation format type of a received signal. In some aspects, such biasing algorithms and apparatus are used in mobile telephone systems implementing EDGE technology, which transmit and receive both GMSK and 8PSK signals. In some aspects, such biasing algorithms and apparatus affect/bias modulation detection of a current burst in a data block of a received signal by detecting/estimating a burst modulation format (GMSK or 8PSK) of the data block using collected information and/or results determined during prior burst modulation detection of one or more of the data blocks. In some aspects, the information collected during modulation detection of a previous burst comprises either a signal-to-noise ratio (SNR) of the previous burst or noise energy information.

Thus, modulation detection information and/or results regarding a burst of a data block are used to bias and improve modulation detection accuracy of one or more subsequent bursts of the same data block. In some aspects, the biasing algorithms and apparatus do not determine the modulation format type of some bursts of a data block in an independent manner (i.e., instead of not considering the detection of other bursts in the same data block when detecting the modulation format of some bursts of a data block, biasing with the detection of one or more other bursts).

In some aspects, the biasing algorithm is implemented by software and/or hardware configured to implement the algorithm. In some aspects, the configured software and/or hardware is installed on a mobile terminal (e.g., a cellular telephone) that is capable of receiving radio signals modulated with at least two different modulation format types (e.g., GMSK and 8 PSK).

Test results show that the modulation detection accuracy of the bias detection algorithm, especially at low signal levels, is improved over conventional detection methods.

Drawings

FIG. 1 is a block diagram of a mobile communication system;

FIG. 2 is a schematic diagram of various components used in a mobile communication system;

FIG. 3 shows a schematic representation of a signal comprising blocks of data, each block comprising four bursts of data;

fig. 4 illustrates SNR statistics for detection and false detection in the case of GMSK modulated signals detected by conventional methods;

fig. 5 illustrates SNR statistics for each data block error detection in the case of conventional detection of GMSK modulated signals;

FIGS. 6A-B are flow diagrams of an improved biasing method for detecting the type of modulation format of a received signal burst using the SNR value of the burst;

FIGS. 7A-B are flow diagrams of an improved biasing method for detecting a type of modulation format of a received signal burst using a burst noise energy value;

FIG. 8 is a flow chart of a method of detecting a data block burst modulation format type based on SNR;

FIG. 9 is a flow chart of a method of detecting a data block burst modulation format type based on noise energy;

FIG. 10 compares the false detection rates of conventional and offset modulation detection methods for detecting GMSK modulated signals;

FIG. 11 compares the error detection rates of conventional and offset modulation detection methods for detecting 8PSK modulated signals;

FIG. 12 compares the total Bit Error Rate (BER) of a GMSK modulated signal detected by the conventional and offset modulation detection methods;

FIG. 13 illustrates a computer system implementing some embodiments.

Detailed Description

In the following description, numerous details are set forth for purposes of explanation. However, it will be apparent to one of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to avoid obscuring the present invention. The word "exemplary" is used herein to mean "serving as an example, instance, or for illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

The following description is divided into four sections. Section I discusses basic terms and concepts related to mobile communications. Section II discusses statistics and test results of conventional modulation detection methods and conclusions that can be drawn from these statistics and test results. Section III discusses improved biasing methods that enable more accurate detection of the type of modulation of the received signal burst. Section IV discusses the improved test results of the bias method when detecting the type of modulation of the received signal burst.

Part I: terms and concepts

Fig. 1 is a block diagram of a mobile communication system 100. The mobile communications system 100 includes one or more base station subsystems 110, a network and switch subsystem 130, one or more mobile terminals 150, and a public switched telephone network 160. The base station subsystem 110 is coupled to a network and switch subsystem 130 and the public switched telephone network 160 communicates with the mobile terminals 150 via over-the-air radio transmissions.

Each base station subsystem 110 typically includes a base station controller 115 and one or more base transceiver stations 120. The base transceiver stations 120 are used to transmit signals to and receive signals from the mobile terminals 150 and include equipment for such operations (e.g., radio transmission towers, etc.). The base station controller 115 is used to communicate signals to the mobile switching center 145 of the network and switch subsystem 130.

The network and switch subsystem 130 typically includes a plurality of local and visitor databases 135, a plurality of authentication centers 140, and a plurality of mobile switching centers 145. The home and visitor location databases 135 are used to store registered user information records, mobile terminal 150 location information, and other information. The authentication center 140 is used to authenticate in conjunction with the home and visitor location databases 135 for security purposes. The mobile switching center 145 is used to hand off signal connections for the public switched telephone network 160 and the base station controller 115.

A registered user of a registered network is able to communicate with other registered users or non-registered users outside the network, such as users within the public switched telephone network 160, using a mobile terminal 150, which mobile terminal 150 comprises receiving means (e.g. a cellular phone, a Personal Digital Assistant (PDA), a laptop, a Blackberry)^TMPersonal Digital Assistants (PDAs), or any other portable computer, etc.).

In some embodiments, the mobile communication system 100 utilizes EDGE technology to transmit and receive both GMSK and 8PSK modulated signals. In these embodiments, hardware or software implemented in various components of the mobile communication system 100 (e.g., the base station controller 115, the base transceiver station 120, the mobile terminal 150, etc.) are configured to transmit and receive both GMSK and 8PSK signals.

Fig. 2 is a schematic diagram illustrating various components used to modulate/encode and demodulate/decode signals in a mobile communication system. The functions of the various components described below are implemented by hardware and/or software configured to modulate/encode and demodulate/decode using GMSK and 8PSK modulation formats.

The various components of fig. 2 include a transmit component 205 for modulating/encoding and transmitting signals, and a receive component 250 for receiving and demodulating/decoding signals. In some embodiments, the transmitting component 205 is a component of a transceiver at a base station. In some embodiments, receiving component 250 is a component of a mobile terminal (e.g., a cellular telephone). The various components of a mobile communication system are well known in the art and will not be described in detail herein.

Transmit component 205 includes a source encoder 210, a channel encoder 215, an interleaver 220, a block divider 225, a modulator 230, and a transmitter 235. The source encoder 210 receives an information bit stream 202 representing the original information to be transmitted. Such raw information may include, for example, voice, music or other audio streams, video streams, email data, video or audio streams or other real-time data, file download operations (e.g., pursuant to a File Transfer Protocol (FTP)), and so forth. However, the above list is not exhaustive, and the information bit stream 202 can also represent other types of original information.

The source encoder 210 compresses the information bit stream and the channel encoder 215 (e.g., a convolutional encoder) is used to introduce redundant bits into the bit stream for error detection and correction at the end of the receiving end. Interleaver 220 interleaves the bits of the information bit stream by swapping the positions of the bits to mitigate the effects of deep fading (in a fading environment). It is noted that the Modulation Coding Scheme (MCS) supporting the 8PSK modulation format employs a different channel encoder 215 and interleaver 220 than the modulation coding scheme supporting the GMSK modulation format.

The information bit stream comprises a plurality of data blocks, which are the basic processing units of the information bit stream. The block divider 225 divides each data block of the information bit stream into four data bursts, while the modulator 230 modulates/encodes the information bit stream by modulating/encoding each burst of the information bit stream into bursts in GMSK or 8PSK format. In general, a modulation format has a certain number of possible symbols to represent bits/to encode bits. A symbol is a possible state (e.g., phase offset) allowed under a particular modulation format. In GMSK format, only two possible symbols/states are used to represent data, where each state represents a single bit. While in the 8PSK format, there are eight possible symbols/states used to represent the data, so each state represents three bits of data. Thus, in GMSK format, one symbol represents one bit of the information bit stream, and in 8PSK format, one symbol represents three bits of the information bit stream. In addition to the information symbols (representing the original information to be transmitted), the GMSK or 8PSK formatted burst also includes Training Sequence (TSC) symbols (as described below in connection with fig. 3) that are used to detect the burst modulation format.

In addition to modulating the information bit stream, modulator 230 also applies the bit stream to a carrier frequency signal, producing a modulated signal for transmission. Modulator 230 also performs other processing steps such as removing intersymbol interference (ISI) components with gaussian filtering, etc. The modulated signal is then transmitted onto a channel by a transmitter 235 using a particular transmission technique, such as time division multiple access.

The receive component 250 includes a receiver 252, a demodulation filter 255, a modulation detector 260, an equalizer 265, a block combiner 270, a deinterleaver 275, a channel decoder 280, a Cyclic Redundancy Check (CRC)282, and a source decoder 285. Receiver 252 receives the modulated signal and demodulation filter 255 removes the carrier frequency signal, producing a "soft decision" signal/output at the baseband frequency. As known to those skilled in the art, a "soft decision" signal/output is not a bit stream with only 0 or 1 values, but includes intermediate values, such as 1.3, 2.5, etc.

The modulation detector 260 estimates the modulation format of the signal burst (GMSK or 8PSK burst) received from the demodulation filter 255. In some embodiments, the modulation detector 260 is implemented by hardware and/or software, where the hardware and/or software is configured to perform the improved bias detection method. After estimating the burst as a GMSK or 8PSK burst, the remaining receiving components 250 treat this burst as one of this format.

Equalizer 265 then removes intersymbol interference (ISI) from the bursts received from modulation detector 260 and block combiner 270 combines the bursts back into data blocks (four bursts per block). Deinterleaver 275 then performs the inverse of the operation of interleaver 220 to rearrange the bits into the positions they were used for. The channel decoder 280 (e.g., a viterbi decoder) then utilizes the redundant bits introduced by the channel encoder 215 for error detection and correction. The CRC282 determines whether the error correction of the burst by the channel decoder 280 is sufficient so that the bits of the burst have been correctly received. If the burst passes the CRC282 determination, it is passed to the source decoder 285, otherwise it is discarded. The source decoder 285 then decompresses the signal received from the channel decoder 280, taking out an information signal 290 (in the form of "soft decisions") that represents the original information (e.g., audio or video stream, e-mail data, etc.) designated for transmission.

After the modulation detector 260 has estimated the burst as a GMSK or 8PSK burst, the remaining receiving components 250, which are to be processed differently depending on this estimation, process this burst as such. For example, the equalizer 265, the deinterleaver 275, and the channel decoder 280 include different devices or perform different processes according to the estimation of the burst as a GMSK or 8PSK burst. As described above, the channel encoder 215 and the interleaver 220, which are used by a Modulation Coding Scheme (MCS) supporting the 8PSK modulation format, are different from those used by a modulation coding scheme supporting the GMSK modulation format. Thus, the deinterleaver 275 and the channel decoder 280 comprise different means depending on the modulation estimate of the burst. If the burst is detected incorrectly and processed according to an incorrect modulation format, the burst is unlikely to pass the CRC282 determination and is discarded.

As described above, signals are typically modulated and transmitted in data blocks (elementary units used to transmit data), where a data block comprises four data bursts. Fig. 3 shows a schematic diagram of a signal comprising data blocks 305, each block 305 comprising four data bursts 310. Conceptually, the burst 310 has a structure including symbols of a specific type arranged in a specific order. The structure of burst 310 generally includes a first set of guard symbols 315 at a first end of the structure, followed by a first set of information symbols 320, a Training Sequence (TSC) symbol 325 at the center of the structure, followed by a second set of information symbols 320, and a second set of guard symbols 315 at a second end of the structure. The first and second sets of information symbols represent original information (e.g., audio or video streams, e-mail data, etc.) that is designated to be transmitted.

TSC symbols 325 generally include symbols used to estimate the channel, frequency error, timing, and modulation format type. In some embodiments, the TSC symbols of a burst are used to determine the burst modulation format type. In these embodiments, the TSC symbols of a burst are used to determine a signal-to-noise ratio (SNR) estimate for the burst as an 8PSK burst and for the burst as a GMSK burst. The modulation format type of the burst is then estimated based on a comparison of the signal-to-noise ratio estimates for both 8PSK and GMSK bursts. In other embodiments, the noise energy estimate of the burst is used instead of the SNR estimate to detect the modulation format type of the burst. These embodiments are described later in connection with fig. 8 and 9.

Section II: test results of conventional detection methods

Fig. 4 shows statistical test results for a conventional modulation detection method for detecting GMSK modulated signals, under typical urban conditions of 50 kilometers per hour (TU50) channel conditions.

As described above, conventional modulation detection determines the modulation type separately from other bursts of the same data block by comparing SNR estimates obtained by treating the bursts as 8PSK and GMSK bursts, respectively. The SNR estimate obtained as an 8PSK burst is sometimes referred to herein as "SNR _8 PSK", and the SNR estimate obtained as a GMSK burst is sometimes referred to herein as "SNR _ GMSK". The absolute value of the difference between these SNR estimates is sometimes referred to herein as "Δ SNR" and can be represented by the following equation: Δ SNR equals abs (SNR _8PSK-SNR _ GMSK).

The upper graph 405 includes a first segmentation line 410 that illustrates the average Δ SNR value (in decibels) for a correct detection example versus the signal SNR. The upper graph 405 also includes a second segment line 415 that illustrates the average Δ SNR value versus signal SNR for the error detection example. The lower graph 420 includes a third segmented line 425 illustrating the Δ SNR standard deviation of a correctly detected example as a function of signal SNR. The lower graph 420 also includes a fourth segmented line 430 that illustrates the Δ SNR standard deviation of the error detection example versus the signal SNR.

Graphs 405 and 420 illustrate that the value of Δ SNR is typically much lower with false detection than with correct detection when conventional modulation detection methods are employed. This is because Δ SNR is equal to the absolute value of the difference between the SNR values of the bursts obtained respectively as 8PSK bursts and as GMSK bursts. When the absolute value of the difference between these SNR values is small, this indicates that the burst is not very clearly an 8PSK or GMSK burst. Therefore, it is indicated that the modulation detection confidence is low when the Δ SNR value is low, and the false detection probability is high in the case where the signal intensity level is low (where the Δ SNR value is small).

Fig. 5 is a graph 500 illustrating further statistical test results for detecting GMSK modulated signals using conventional modulation detection methods under TU50 channel conditions. The diagram 500 includes a first segmentation line 505 that illustrates the number of data blocks in a block for which a burst is error free to detect, as a function of the SNR of the GMSK signal. The diagram 500 also includes a second segmented line 510 that illustrates the number of data blocks in a block that have one error detection burst as a function of the SNR of the GMSK signal. The third, fourth and fifth segment lines 515, 520 and 525 illustrate the number of data blocks in a block with two, three and four error detection bursts, respectively, as a function of the SNR of the GMSK signal. Graph 500 illustrates that with conventional modulation detection methods, the probability of more than one burst per data block having false detections is low.

Therefore, statistical tests of conventional modulation detection methods show that there is a high probability of false detection in the case of low signal strength levels (where Δ SNR values are small), and that there is a low probability of false detection of more than one burst per data block.

Section III: improved bias modulation detection

In some embodiments, the statistical characteristics of the false detection cases of conventional detection techniques (as described in section II) are utilized to derive an improved bias modulation detection method. In some embodiments, such biasing methods utilize information and/or results collected during modulation detection of one or more previous bursts of a data block to affect/bias modulation detection of a current burst of the same data block. In some embodiments, the information collected during modulation detection of a previous burst includes SNR or noise energy information of the previous burst. Thus, modulation detection information and/or results regarding a burst of data blocks are used to bias and improve modulation detection accuracy for one or more subsequent bursts of the same data block. In some embodiments, this biasing method cannot determine the modulation format of the burst of data blocks in an independent manner.

Fig. 6A-B are flow diagrams of an improved biasing method 600 for detecting a burst modulation format type of a received signal using SNR values of the bursts. In some embodiments, software and/or hardware is configured to implement the method 600. In some embodiments, configured software and/or hardware is installed on a receiving apparatus, such as a mobile terminal, that receives modulated radio signals of at least two modulation format types (e.g., GMSK and 8 PSK).

The method 600 begins with receiving (at 605) an allocated (from a base station) TSC symbol and a signal that has been modulated with one of two or more different modulation format types. In the embodiments described below, the received signal is modulated with either a first modulation format type, i.e., GMSK, or a second modulation format type, i.e., 8 PSK. In other embodiments, the received signal is modulated with a different format type and/or a different number of format types. The received modulated signal includes a plurality of data blocks, each data block including a plurality of bursts. In the embodiments described below, the data block comprises four data bursts. In other embodiments, the data blocks have different numbers of bursts.

The modulated signal and the assigned TSC symbols are typically received from a base station that allocates the resources required for multiple simultaneous modulated signals (telephone calls). In allocating resources, the base station allocates one of typically seven possible configurations of TSC symbols to each modulated signal. In some embodiments, the assigned TSC symbols are associated with the TSC symbols of the burst to help determine the modulation format type of the burst (as described below with reference to fig. 8).

The method 600 next sets (at 610) the data block of the received signal to the current data block to be processed. Next the method estimates/detects (at 615) the modulation format type of the first burst of the current data block (as GMSK or 8PSK burst) using an SNR detection method that determines an SNR estimate obtained using the first burst as GMSK burst and as 8PSK burst. One such SNR detection method is described below with reference to fig. 8, although other SNR detection methods are employed in other embodiments. The method next processes (at 617) the first burst according to the estimate (i.e., decodes/demodulates the burst as GMSK or 8PSK using suitable software and/or hardware configured to accomplish this task).

The method then calculates (at 620) the difference in SNR estimates obtained by treating the first burst as a GMSK burst and as a 8PSK burst (denoted as the first Δ SNR or Δ SNR)₁). The method 600 then performs modulation format type estimation/detection (at 625) of the second burst of the current data block (as a GMSK or 8PSK burst) using an SNR detection method that determines an SNR estimate of the second burst as GMSK or 8PSK burst. The method then calculates (at 630) the difference in the SNR estimates obtained by treating the second burst as a GMSK burst and as an 8PSK burst (denoted as the second Δ SNR or Δ SNR)₂). The method then uses the collected information (Δ SNR)₁) And biasing the estimate of the modulation format type of the second burst with the estimate of the modulation detection of the first burst. In some embodiments, the method further utilizes a first pre-stageThe confidence measure biases (in decibels) the estimate of the second burst.

To bias the estimate of the second burst, the method 600 determines (at 635) whether the modulation format estimate of the second burst is different from the estimate of the first burst. If not, the method proceeds to step 655. However, if the modulation format estimates are different, the method next determines (at 640) whether the difference between the respective first and second Δ SNRs for the first and second bursts is greater than a first predetermined confidence measure. If so, the method resets (at 650) the modulation format estimate for the second burst to be equal to the estimate for the modulation format for the first burst (determined at 615). If not, the method proceeds to step 655, where the method processes the second burst according to the estimate.

In some embodiments, steps 635 through 650 of the method 600 are represented by the following algorithm:

if((Mod_type₂≠Mod_type₁)&(abs(ΔSNR₁)-abs(ΔSNR₂)＞λ₁db))

then(Mod_type₂＝Mod_type₁)

wherein

Mod_type₁A modulation format estimate for the first burst;

Mod_type₂a modulation format estimate for the second burst;

ΔSNR₁a first Δ SNR for the first burst;

ΔSNR₂a first Δ SNR for the second burst; and

λ₁db is the first predetermined confidence measure.

The method then continues with bias estimation for each subsequent burst of the data block. In step 657, the method applies SNR detection to the third burst of the current data blockModulation format estimation (estimated as GMSK or 8PSK bursts) is performed and the SNR detection method determines the SNR estimate for the third burst, which is GMSK and 8PSK bursts. The method next calculates (at 660) the difference between the SNR estimates of the third burst (which will be called third Δ SNR or Δ SNR) as GMSK and 8PSK bursts₃). The method then uses the collected information (Δ SNR)₁And Δ SNR₂) And the estimated results of the first and second burst modulation detections, biasing the modulation format estimate of the third burst. In some embodiments, the method further biases the estimate of the third burst with a second predetermined confidence measure (in decibels).

To bias the estimate of the third burst, the method 600 determines (at 665) whether the estimate of the third burst is different from the estimate of the first burst and different from the estimate of the second burst. If not, the method proceeds to step 680. However, if the condition is true, then the method next determines (at 670) (Δ SNR)₁And Δ SNR₂) Average of and Δ SNR₃Whether the difference between is greater than a second predetermined confidence measure. If so, the method resets (at 675) the estimate of the third burst to be equal to the estimate of the first burst modulation format (determined at step 615). If not, the method proceeds to step 680 where the method processes a third burst based on the estimate.

In some embodiments, steps 665 through 675 of the method 600 are represented by the following algorithm:

if((Mod_type₃≠Mod_type₂)&(Mod_type₃≠Mod_type₁)&(abs(Mean(ΔSNR₁，ΔSNR₂))-abs(ΔSNR₃)＞λ₂db))then(Mod_type₃＝Mod_type₁)

wherein Mod _ type₃A modulation format estimate for the third burst;

ΔSNR₃first Δ SNR for the third burst; and is

λ₂db is the second predetermined confidence measure.

At step 681, the method performs modulation format estimation (GMSK or 8PSK bursts) on the fourth burst of the current data block using SNR detection methods or other detection methods known in the art that are not based on SNR. The method then biases the modulation format estimate for the fourth burst with the estimates of the first, second, and third bursts. In some embodiments, if the estimates of the first, second and third bursts are all the same (all estimates of the same modulation format), then the estimate of the fourth burst is reset to be equal to the estimates of the other bursts (so that all four bursts have the same modulation estimate).

To bias the estimate of the fourth burst, the method 600 determines (at 682) whether the estimate of the fourth burst is different from the estimates of the first, second, and third bursts. If so, the method resets (at 685) the estimate of the fourth burst to be equal to the estimate of the first burst modulation format (determined at step 615). If not, the method proceeds to step 688, where the method processes a fourth burst according to the estimate.

In some embodiments, steps 682 through 685 of the method 600 are represented by the following algorithm:

if((Mod_type₄≠Mod_type₃)&(Mod_type₄≠Mod_type₂)&(Mod_type₄≠Mod_type₁))

then(Mod_type₄＝Mod_type₁)

wherein Mod _ type₄The modulation format estimate for the fourth burst.

The method determines whether there are more data blocks in the received signal to process in step 690. If so, the method proceeds to step 610, where the next data block in the received signal is set to the current data block to be processed. If not, the method ends.

In some embodiments, the predetermined confidence measure (λ) is determined by experimentally finding the confidence measure value that results in the best detection accuracy for method 600₁And λ₂). In some embodiments, the predetermined confidence measure (λ)₁And λ₂) Are equal in value. In other embodiments, the confidence measure (λ) is predetermined₁And λ₂) Are not equal in value. In some embodiments, the predetermined confidence measure (λ)₁And λ₂) The values range from 0 to 4 db.

As mentioned above, a low Δ SNR for burst modulation detection indicates a low confidence in this detection, since this indicates that the burst is not very clearly an 8PSK burst or a GMSK burst. Thus, if the Δ SNR of the previous burst is significantly greater than the Δ SNR of the current burst, indicating that the confidence in the detection of the previous burst is significantly greater than the confidence in the detection of the current burst, method 600 biases the detection of the current burst to be equal to the detection of the previous burst (if the two detections are not already equal). Conceptually, the value of the predetermined confidence measure determines what "significantly greater than" means, and sets the threshold level when the bias of method 600 is in effect. The lower the value of the predetermined confidence measure, the greater the biasing effect of the detection method 600.

Note also that method 600 determines whether the estimate of the subsequent burst is different from the estimates of one or more preceding bursts, and if so, may bias the estimate of the subsequent burst. This reflects the statistical observation that there is a low probability of more than one false detection in a data block (as described above). Thus, the method 600 utilizes two statistical observations of conventional detection methods to provide a more accurate modulation detection method, as discussed below in section IV.

Fig. 7A-B are flow diagrams of an improved biasing method 700 for detecting a modulation format type of a received signal burst using a noise energy value of the burst. In some embodiments, software and/or hardware is configured to implement method 700. In some embodiments, configured software and/or hardware is installed on a receiving apparatus that receives radio signals (such as a mobile terminal) modulated with at least two different modulation format types (e.g., GMSK and 8 PSK).

The noise energy bias method 700 includes steps similar to those of the SNR bias method 600 shown in fig. 6 and will not be described in detail herein. However, method 700 achieves detection purposes with noise energy values rather than SNR values, and therefore, certain steps of method 600 are modified to reflect the inverse relationship between burst SNR levels and burst noise energy levels (SNR ═ channel energy/noise energy). Here, the Noise Energy (NE) estimate of a burst corresponding to 8PSK may be referred to as "NE _8 PSK" and the Noise Energy (NE) estimate of a burst corresponding to GMSK may be referred to as "NE _ GMSK". The absolute value of the difference between these NE estimates, sometimes referred to herein as "Δ NE", can be expressed by the following equation: Δ NE ═ abs (NE _8PSK-NE _ GMSK).

The method 700 begins when an assigned TSC symbol and a signal that has been modulated with one of two or more different modulation format types are received (at 705) from a base station. In some embodiments, the assigned TSC symbols are correlated with the TSC symbols of the burst to help determine the modulation format type of the burst (as discussed later with reference to fig. 9).

The method 700 next sets (at 710) the data block of the received signal to the data block currently to be processed. The method next estimates/detects (at 715) the modulation format type (GMSK or 8PSK burst) of the first burst of the current data block using a noise energy detection method that determines the first burst as a GMSK burst and as a noise energy estimate for the 8PSK burst. One such noise energy detection method is described below with reference to fig. 9, although other noise energy detection methods are employed in other embodiments. The method next processes the first burst according to the estimate (at 717).

Method 700 next calculates (at 720) the difference between the noise energy estimates obtained by treating the first burst as a GMSK burst and as a 8PSK burst (note thatAs first Δ NE or Δ NE₁). The method 700 then performs an initial estimation/detection of the modulation format type for the second burst of the current data block using a noise energy detection method (at 725) that determines that the second burst is to be considered as a noise energy estimate for GMSK and 8PSK bursts. The method then calculates (at 730) the difference (denoted as second Δ NE or Δ NE) between the noise energy estimates obtained with the second burst as GMSK burst and with the 8PSK burst₂). The method then uses the collected information (Δ NE)₁) And an estimate of the modulation detection for the first burst, biasing the modulation format type estimate for the second burst. In some embodiments, the method further biases the estimate of the second burst with a first predetermined confidence measure (in watts).

To bias the estimate of the second burst, the method 700 first determines (at 735) whether the modulation format estimate of the second burst is different from the estimate of the first burst. If not, the method proceeds to step 755. If, however, the modulation format estimates are different, the method next determines (at 740) whether the difference between the first and second Δ NE of the first and second bursts, respectively, is greater than a first predetermined confidence measure (as described below, since a low Δ NE value means that the confidence in the modulation detection is low). If so, the method resets (at 750) the modulation format estimate for the second burst to be equal to the modulation format estimate for the first burst (determined at 715). If not, the method proceeds to step 755, where the method processes the second burst according to the estimate.

In some embodiments, steps 735 through 750 of the method 700 are represented by the following algorithm:

if((Mod_type₂≠Mod_type₁)&(abs(ΔNE₁)-abs(ΔNE₂)＞λ₁))

then(Mod_type₂＝Mod_type₁)

wherein Mod _ type₁A modulation format estimate for the first burst;

Mod_type₂a modulation format estimate for the second burst;

ΔNE₁first Δ NE of the first burst;

ΔNE₂first Δ NE of the second burst; and

λ₁watt is the first predetermined confidence measure.

At step 757, the method performs an initial estimate of the modulation format (GMSK or 8PSK burst) on the third burst of the current data block using a noise energy detection method that determines the third burst as an estimate of the noise energy of the GMSK and 8PSK bursts. The method then calculates (at 760) a third burst as the difference between the noise energy estimates of the GMSK and 8PSK bursts (denoted as third Δ NE or Δ NE)₃). The method then uses the collected information (Δ NE)₁And Δ NE₂) And biasing the modulation format estimate for the third burst as a result of the estimation of the modulation detections for the first and second bursts. In some embodiments, the method further biases the estimate of the third burst with a second predetermined confidence measure.

To bias the estimate of the third burst, the method 700 first determines (at 765) whether the estimate of the third burst is different from the estimate of the first burst and different from the estimate of the second burst. If not, the method proceeds to step 780. However, if the condition is true, then the method next determines (at 770) (Δ NE)₁And Δ NE₂) Average of and Δ NE₃Whether the difference is greater than a second predetermined confidence measure. If so, the method resets (at 775) the estimate of the third burst to be equal to the modulation format estimate of the first burst (determined at 715). If not, the method proceeds to step 780 where the method processes a third burst based on the estimate.

In some embodiments, steps 765 through 775 of the method 700 are represented by the following algorithm:

if((Mod_type₃≠Mod_type₂)&(Mod_type₃≠Mod_type₁)&(abs(Mean(ΔNE₁，ΔNE₂))-abs(ΔNE₃)＞λ₂)

then(Mod_type₃＝Mod_type₁)

wherein Mod _ type₃A modulation format estimate for the third burst;

ΔNE₃first Δ NE of the third burst; and

λ₂watt is a second predetermined confidence measure.

At step 781, the method performs an initial estimation of the modulation format for the fourth burst of the current data block using a noise energy detection method or other detection methods known in the art that are not based on noise energy. The method then biases the modulation format estimate for the fourth burst with the estimates of the first, second, and third bursts. In some embodiments, if the estimates of the first, second and third bursts are all the same (i.e., are all estimates of the same modulation format), then the estimate of the fourth burst is reset to be equal to the estimates of the other bursts (so all four bursts have the same modulation estimate).

To bias the estimate of the fourth burst, the method 700 determines (at 782) whether the estimate of the fourth burst is different from the estimates of the first, second, and third bursts. If so, the method resets (at 785) the estimate of the fourth burst to be equal to the estimate of the first burst modulation format (determined at 715). If not, the method proceeds to step 788, where the method processes a fourth burst based on the estimate.

In some embodiments, steps 782 through 785 of the method 700 are represented by the following algorithm:

if((Mod_type₄≠Mod_type₃)&(Mod_type₄≠Mod_type₂)&(Mod_type₄≠Mod_type₁))

then(Mod_type₄＝Mod_type₁)

wherein Mod _ type₄The modulation format estimate for the fourth burst.

At step 790, the method determines if there are more data blocks in the received signal to process. If so, the method proceeds to step 710 where the next data block in the received signal is set to the current data block to be processed. If not, the method ends.

In some embodiments, the predetermined confidence measure (λ)₁And λ₂) Is determined by experimentation to find the confidence measure that results in the best detection accuracy for method 700. In some embodiments, the predetermined confidence measure (λ)₁And λ₂) Are equal in value. In other embodiments, the confidence measure (λ) is predetermined₁And λ₂) Are not equal in value. In some embodiments, the predetermined confidence measure (λ)₁And λ₂) The value ranges from 0 to 4 watts.

Burst modulation detection with low Δ SNR indicates low confidence in the detection, since this indicates that the burst is not clearly an 8PSK burst or a GMSK burst. Thus, if Δ NE of the previous burst is significantly greater than Δ NE of the current burst, indicating that the confidence in the detection of the previous burst is significantly greater than the confidence in the detection of the current burst, the method 700 biases the detection of the current burst to be equal to the detection of the previous burst (if the two detections are not already equal). Conceptually, the predetermined confidence measure determines what "significantly greater than" means, and sets the threshold level when the bias of method 700 is in effect. The lower the value of the predetermined confidence measure, the greater the biasing effect of the detection method 700.

Fig. 8 is a flow chart of a SNR-based method 800 for determining/detecting a modulation format type of a data block burst. In some embodiments, method 800 includes step 615 of method 600 (illustrated with reference to fig. 6) in which a modulated signal and assigned TSC symbols have been received (at step 605). In some embodiments, method 800 also includes steps 625, 657, and 681 of method 600.

In summary, the method 800 utilizes the assigned TSC symbols and the TSC symbols of the bursts of the received signal to determine the bursts as 8PSK bursts and as SNR estimates for the GMSK bursts. The modulation format type of the burst is then estimated from a comparison of the SNR estimates of the burst (as 8PSK and GMSK bursts).

The method starts by treating this burst as a GMSK burst and calculating (at 805) the SNR (SNR _ GMSK) of the burst as GMSK. The method may do so, for example, by correlating (multiplying) the TSC symbols of the burst (interpreted as GMSK symbols) with the assigned TSC symbols, resulting in a TSC symbol product. The allocated TSC symbols are configured such that: if the TSC symbols of this burst are correctly interpreted as GMSK symbols (i.e. in fact this burst is a GMSK burst), the resulting TSC symbol product should have a high correlation value (spike) in the middle of the TSC symbol product and a small value (close to zero) at the left and right ends of the TSC symbol product. Conversely, if the TSC symbols of this burst are erroneously interpreted as GMSK symbols (i.e., in fact the burst is not GMSK), there is no significant correlation spike in the middle of the TSC symbol product.

The correlation spike in the middle of the TSC symbol product can then be used to determine the channel energy (signal strength) and noise energy of this burst. This is done by estimating the channel taps based on the correlation peaks, where the sum of the channel tap amplitudes gives us the total channel energy. If it is notIs the estimated channel tap width (typically between 4-9 channel taps, each tap being spaced by a particular symbol duration depending on the channel model for which the system is designed), then the channel energyThe amount is equal to:

the estimated channel taps and assigned TSC symbols are used to reconstruct the received signal and determine the estimated error energy (i.e., the error energy of the reconstructed signal compared to the actual received signal in the TSC region). The noise energy of the burst can then be estimated on the TSC symbols by taking the mean square of this estimation error. Thus, the noise energy of the burst is equal to:

wherein:

x_ian allocated TSC symbol;

y_ia TSC code element of the pulse string;

n — the length of the TSC (e.g., 26 symbols long for GSM and EDGE); and

j is the number of channel taps in the channel estimate.

The SNR of this burst (as GMSK burst) can then be calculated from the channel energy and the noise energy, expressed as the following equation: SNR is channel energy/noise energy.

The method 800 next calculates (at 810) the SNR (SNR _8PSK) for this burst to be treated as an 8PSK burst. This method may do so by repeating the steps described above with respect to step 805, except that the method treats the burst as an 8PSK burst. The method next determines (at 815) whether the SNR estimate for the GMSK burst is greater than the SNR estimate for the 8PSK burst. If so, then the method 800 determines/estimates (at 820) that this burst is modulated as a GMSK burst (since a higher SNR value for this burst as a GMSK burst indicates that the burst is likely a GMSK burst). If not, then method 800 determines/estimates (at 825) that the burst is modulated to 8PSK burst (since a higher SNR value for the burst as an 8PSK burst indicates that the burst is likely to be an 8PSK burst). The method then ends.

Fig. 9 is a flow chart of a method 900 for determining/detecting a modulation format type of a burst of a data block based on noise energy. In some embodiments, method 900 includes a step 715 of method 700 (described with reference to fig. 7), where method 700 has received (at step 705) the modulated signal and the assigned TSC symbols. In some embodiments, method 900 also includes steps 725, 757, and 781 of method 700. Method 900 includes steps similar to those of method 800 of fig. 8 (except that the noise energy value of the burst is substituted for the SNR value of the burst) and will not be described in detail herein.

The method starts by treating the burst as a GMSK burst and calculating (at 905) the noise energy (NE _ GMSK) of this burst as GMSK burst. This method may do so, for example, by correlating the TSC symbols of a burst (interpreted as GMSK symbols) with the assigned TSC symbols, resulting in a TSC symbol product. The correlation spike of the TSC symbol product is then used to determine the noise energy of the burst (as described above).

Method 900 next calculates (at 910) the burst as the noise energy (NE _8PSK) of an 8PSK burst. In addition to treating the burst as an 8PSK burst, the method may do so by repeating the steps described above for step 905. The method next determines (at 915) whether the noise energy estimate for the GMSK burst is less than the noise energy estimate for the 8PSK burst. If so, then the method 900 determines/estimates (at 920) that the burst is modulated as a GMSK burst (since the noise energy value of the burst as a GMSK burst is low indicating that the burst is likely to be a GMSK burst). If not, then the method 900 determines/estimates (at 925) that this burst is modulated to an 8PSK burst (since the lower noise energy value of this burst as an 8PSK burst indicates that this burst is likely to be an 8PSK burst). The method then ends.

Section IV: test results of improved detection methods

Fig. 10 shows a graph 1000 comparing the false detection rate of conventional and offset modulation detection methods for detecting GMSK modulated signals under TU50 channel conditions for the 1900MHz band. The diagram 1000 includes a first segmented line 1005 illustrating the variation of the detection block error rate of the offset detection method with the SNR of the GMSK signal. The diagram 1000 also includes a second segmented line 1010 illustrating the variation of the detection block error rate of conventional detection methods with the SNR of the GMSK signal.

Fig. 11 shows a graph 1100, graph 1100 comparing the error detection rate of a conventional and offset modulation detection method for detecting an 8PSK modulated signal under channel conditions of TU50 for the 1900MHz band. Diagram 1100 includes a first segmented line 1105 that illustrates the variation of the detection block error rate of this offset detection method with the SNR of an 8PSK signal. Diagram 1100 also includes a second segmented line 1110 that illustrates the variation of the detection block error rate with 8PSK signal SNR for conventional detection methods.

Fig. 10 and 11 illustrate that the offset modulation detection method is about 3-5 db improved over the conventional modulation detection method for both GMSK and 8PSK (i.e., the offset detection method has the same detection error rate as the conventional detection method when detecting at a 3-5 db lower signal level than the conventional detection method).

Fig. 12 shows a graph 1200 that compares the total Bit Error Rate (BER) performance of conventional and offset modulation detection methods for detecting GMSK modulated signals under TU50 channel conditions for the 1900MHz band. Diagram 1200 includes a first segmented line 1205 that illustrates the variation of the detected bit error rate of this bias detection method as a function of the GMSK signal SNR. The diagram 1200 also includes a second segmented line 1210 that illustrates the variation of the detected bit error rate of this conventional detection method as a function of the SNR of the GMSK signal. FIG. 12 illustrates the bit error rate improvement of this offset detection method of 0.25-0.5 db at a 10% BER point over the conventional detection method. This improved detection performance also results in an improved block error rate (BLER).

As shown in fig. 10, 12 and 12, the test results demonstrate that this offset detection method provides an improvement in modulation detection accuracy over conventional detection methods.

Fig. 13 illustrates a computer system 1300 with which some embodiments are implemented. In some embodiments, computer system 1300 includes a receiving device (mobile terminal). Computer system 1300 includes a bus 1305, a processor 1310, a system memory 1315, a read only memory 1320, a persistent storage device 1325, an input device 1330, and an output device 1335.

Bus 1305 collectively represents the overall system, peripheral devices, and chipset bus that connects the many internal devices of computer system 1300 for communication. For example, bus 1305 can communicatively connect processor 1310 with read only memory 1320, system memory 1315, and permanent storage 1325.

Read Only Memory (ROM)1320 stores static data and instructions for processor 1310 and other computer system modules. Persistent storage 1325, on the other hand, is a read-write memory device. This device is a non-volatile memory unit that can hold instructions and data even when the computer system 1300 is turned off. Some embodiments utilize mass storage (such as magnetic or optical disks and their corresponding disk drives) as the persistent storage 1325. Other embodiments utilize removable storage devices (such as floppy disks or zip ® disks and their corresponding disk drives) as the permanent storage device.

Like the persistent storage 1325, the system memory 1315 is a read-write memory device. Unlike storage 1325, however, the system memory is a volatile read-and-write memory, such as Random Access Memory (RAM). The system memory holds some of the instructions and data that the processor needs at run-time.

Instructions and/or data needed to perform the methodologies of some embodiments are stored in system memory 1315, persistent storage 1325, read only memory 1320, or any combination of the three. For example, various memory units may include instructions for detecting a received signal modulation format type consistent with some embodiments. From these various memory units, processor 1310 fetches instructions to be executed and data to be processed to perform the processing methods of some embodiments.

The bus 1305 is also connected to input and output devices 1330 and 1335. Input device 1330 allows a user to send information and select commands to computer system 1300. The input device 1330 includes a numeric keypad and a cursor control. Output device 1335 displays images generated by computer system 1300. Output devices include printers and display devices, such as Cathode Ray Tubes (CRTs) or Liquid Crystal Displays (LCDs).

Finally, as shown in FIG. 13, bus 1305 also remotely connects (via a wireless transmission) computer system 1300 to a mobile system 1365, such as through a receiver (not shown). In this manner, the computer system 1300 may be part of the mobile system 1365. Any or all of the components of computer system 1300 may be used in conjunction with some embodiments. However, it will be apparent to those skilled in the art that any other system configuration may be used in combination with other embodiments.

Those of skill in the art would understand that information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

The various illustrative logical blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

The methods or method steps described in connection with the embodiments disclosed herein may be embodied directly in hardware (i.e., in a hard-wired fashion), in a software module executed by a processor, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In addition, such a storage medium may be integrated into the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a mobile terminal. The processor and the storage medium may reside as discrete components in a mobile terminal.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (20)

1. A computer program product comprising a computer readable medium having stored thereon instructions that, when executed, estimate a modulation format of a signal modulated with one of at least two possible modulation formats, the signal comprising a plurality of blocks, each block comprising a plurality of bursts, the computer program product comprising sets of instructions for:

determining a first modulation format estimate for a first burst of blocks; and

determining a second modulation format estimate for the second burst of blocks using information related to the first modulation format estimate.

2. The computer program product of claim 1, wherein the information related to the first modulation format estimate comprises a result of the first modulation format estimate, signal-to-noise ratio (SNR) information for the first burst, or noise energy information for the first burst.

3. The computer program product of claim 1, further comprising sets of instructions for:

determining a third modulation format estimate for the third burst of the block using information related to the first and second modulation format estimates; and

determining a fourth modulation format estimate for the fourth burst of the block using information associated with the first, second, and third modulation format estimates.

4. The computer program product of claim 1, wherein:

the bursts of the signal are modulated with a first or second format;

the set of instructions for determining the second modulation format estimate comprises sets of instructions for:

determining a first signal-to-noise ratio (SNR) delta comprising a difference between signal-to-noise ratios of the first burst modulated with the first format and modulated with the second format;

determining a second signal-to-noise ratio delta comprising a difference between signal-to-noise ratios of said second burst modulated with said first format and said second format;

determining that a difference between the first and second signal-to-noise ratio increments is greater than a predetermined confidence measure;

determining that the second modulation format estimate is not equal to the first modulation format estimate; and

resetting the second modulation format estimate to be equal to the first modulation format estimate.

5. The computer program product of claim 4, wherein the predetermined confidence measure ranges from 0 to 4 decibels.

6. The computer program product of claim 1, wherein:

the bursts of the signal are modulated with a first or second format;

the set of instructions for determining the second modulation format estimate comprises sets of instructions for:

determining a first Noise Energy (NE) delta comprising a difference between noise energy of said first burst modulated with said first format and modulated with said second format;

determining a second noise energy delta comprising a difference between noise energy of said second burst modulated with said first format and modulated with said second format;

determining that a difference between the first and second noise energy increments is greater than a predetermined confidence measure;

determining that the second modulation format estimate is not equal to the first modulation format estimate; and

resetting the second modulation format estimate to be equal to the first modulation format estimate.

7. The computer program product of claim 6, wherein the predetermined confidence measure ranges in value from 0 to 4 watts.

8. The computer program product of claim 1, wherein:

the signals are transmitted and received with a mobile telephone system implementing enhanced data rates for global evolution (EDGE) technology; and

the signal is modulated using a Gaussian Minimum Shift Keying (GMSK) or 8 phase shift keying (8PSK) format.

9. The computer-program product of claim 1, wherein the set of instructions is executed on a mobile terminal receiving the signal.

10. An apparatus configured to estimate a modulation format of a signal, the signal being modulated with one of at least two possible modulation formats, the signal comprising a plurality of blocks, each block comprising a plurality of bursts, the apparatus comprising:

means for determining a first modulation format estimate for a first burst of blocks; and

means for determining a second modulation format estimate for the second burst of the block using information associated with the first modulation format estimate.

11. The apparatus of claim 10, wherein the information related to the first modulation format estimate comprises a result of the first modulation format estimate, signal-to-noise ratio (SNR) information for the first burst, or noise energy information for the first burst.

12. The apparatus of claim 10, further comprising:

means for determining a third modulation format estimate for the third burst of the block using information related to the first and second modulation format estimates; and

means for determining a fourth modulation format estimate for the fourth burst of the block using information associated with the first, second, and third modulation format estimates.

13. The apparatus of claim 10, wherein:

the signals are transmitted and received with a mobile telephone system implementing enhanced data rates for global evolution (EDGE) technology; and

the signal is modulated using a Gaussian Minimum Shift Keying (GMSK) or 8 phase shift keying (8PSK) format.

14. The apparatus of claim 10, wherein the apparatus comprises a mobile terminal that receives the signal.

15. A mobile terminal, comprising:

receiving means configured to receive a signal, said signal being modulated with one of at least two possible modulation formats, said received signal comprising a plurality of blocks, each block comprising a plurality of bursts;

a detection device coupled with the receiving device, the detection device configured to:

determining a first modulation format estimate for a first burst of blocks; and

determining a second modulation format estimate for the second burst of blocks using information related to the first modulation format estimate.

16. The mobile terminal of claim 15, wherein the information related to the first modulation format estimate comprises a result of the first modulation format estimate, signal-to-noise ratio (SNR) information for the first burst, or noise energy information for the first burst.

17. The mobile terminal of claim 15, wherein the detecting means is further configured for:

determining a third modulation format estimate for the third burst of the block using information related to the first and second modulation format estimates; and

determining a fourth modulation format estimate for the fourth burst of the block using information associated with the first, second, and third modulation format estimates.

18. The mobile terminal of claim 15, wherein the detecting means comprises:

a processor; and

memory means, coupled to the processor, for storing instructions for causing the processor to perform the first and second modulation format estimations.

19. The mobile terminal of claim 15, wherein the detecting means comprises hard-wired means for determining the first and second modulation format estimates.

20. The mobile terminal of claim 15, wherein:

the mobile terminal receiving the signal from a base station; and

the mobile terminal and the base station are components in a mobile telephone system implementing enhanced data rates for global evolution (EDGE) technology, which transmits and receives signals modulated with Gaussian Minimum Shift Keying (GMSK) or 8 phase shift keying (8PSK) format.