Burst detector
A burst detector and method for detecting when a selected timeslot which includes data signals for a plurality of channels is received without selected ones of the plurality of channels. The burst detector includes a comparator for generating a burst detection signal when a signal power estimate of a predetermined code detected within a signal received in a selected timeslot is greater than a scaled noise power estimate of any signal received in the selected timeslot. The burst detector also includes a data estimation device for decoding the received signal of the timeslot when the burst detection signal is generated.
Latest InterDigital Technology Corporation Patents:
The application is a continuation of U.S. patent application Ser. No. 10/196,857, filed Jul. 16, 2002, which in turn claims priority from Provisional Patent Application Ser. No. 60/325,692, filed Sep. 28, 2001, which is incorporated by reference as if fully set forth.
BACKGROUNDThe present invention relates to the field of wireless communications. More specifically, the present invention relates to detecting codes in a communication signal in order to activate the receiver to process the signal.
Spread spectrum TDD systems carry multiple communications over the same spectrum. The multiple signals are distinguished by their respective chip code sequences (codes). Referring to
The allocated set of physical channels for each CCTrCh holds the maximum number of RUs that would need to be transmitted during a TTI. The actual number of physical channels that are transmitted during a TTI is signaled to the receiver via the Transport Format Combination Index (TFCI). During normal operation, the first timeslot allocated to a CCTrCh will contain the required physical channels to transmit the RUs and the TFCI. After the receiver demodulates and decodes the TFCI it would know how many RUs are transmitted in a TTI, including those in the first timeslot. The TFCI conveys information about the number of RUs.
DTX can be classified into two categories : 1) partial DTX; and 2) full DTX. During partial DTX, a CCTrCh is active but less than the maximum number of RUs are filled with data and some physical channels are not transmitted. The first timeslot allocated to the CCTrCh will contain at least one physical channel to transmit one RU and the TFCI word, where the TFCI word signals that less than the maximum number of physical channels allocated for the transmission, but greater than zero (0), have been transmitted.
During full DTX, no data is provided to a CCTrCh and therefore, there are no RUs at all to transmit. Special bursts are periodically transmitted during full DTX and identified by a zero (0) valued TFCI in the first physical channel of the first timeslot allocated to the CCTrCh. The first special burst received in a CCTrCh after a normal CCTrCh transmission or a CCTrCh in the partial DTX state indicates the start of full DTX. Subsequent special bursts are transmitted every Special Burst Scheduling Parameter (SBSP) frames, wherein the SBSP is a predetermined interval. Frames 3 and 7 illustrate the CCTrCh comprising this special burst. Frames 4-6 and 8 illustrate frames between special bursts for a CCTrCh in full DTX.
As shown in Frame 9 of
Receivers are able to utilize the receipt of subsequent special bursts to indicate that the CCTrCh is still in the full DTX state. Detection of the special burst, though, does not provide any information as to whether the CCTrCh will be in the partial DTX state or normal transmission state during the next frame.
Support for DTX has implications to several receiver functions, notably code detection. If no codes are sent in the particular CCTrCh in one of its frames, the code detector may declare that multiple codes are present, resulting in a Multi-User Detector (MUD) executing and including codes that were not transmitted, reducing the performance of other CCTrChs that are also processed with the MUD. Reliable detection of full DTX will prevent the declaring of the presence of codes when a CCTrCh is inactive. Also, full DTX detection can result in reduced power dissipation that can be realized by processing only those codes that have been transmitted and not processing empty timeslots.
Accordingly, there exists a need for an improved receiver.
SUMMARYThe present invention is a receiver for receiving a communication signal divided into a plurality of timeslots, wherein the timeslots include a plurality of channels, including a burst detector for detecting when a selected one of the plurality of channels of the communication is received. The burst detector comprises a noise estimation device for determining a scaled noise power estimate of the selected one of the timeslots, a matched filter for detecting signal power of the selected one of the channels of the timeslots and a signal power estimation device, responsive to the matched filter, for generating a signal power estimate of the selected one of the channels of the timeslots. A comparator, responsive to the scaled noise power estimate and the signal power estimate, for generating a burst detection signal when the signal power estimate is greater than the scaled noise power estimate, and a data estimation device, responsive to the burst detection signal, for decoding the plurality of channels are also included in the burst detector.
BRIEF DESCRIPTION OF THE DRAWING(S)
The preferred embodiments will be described with reference to the drawing figures where like numerals represent like elements throughout.
Referring to
The receiver 19 receives various radio frequency (RF) signals including communications over the wireless radio channel using the antenna 5, or alternatively an antenna array. The received signals are passed through a transmit/receive (T/R) switch 6 to a demodulator 8 to produce a baseband signal. The baseband signal is processed, such as by the channel estimation device 7 and the data estimation device 2, in the timeslots and with the appropriate codes assigned to the receiver 19. The channel estimation device 7 commonly uses the training sequence component in the baseband signal to provide channel information, such as channel impulse responses. The channel information is used by the data estimation device 2 and the burst detector 10. The data estimation device 2 recovers data from the channel by estimating soft symbols using the channel information.
The received and demodulated communication is also forwarded to the matched filter 12, as well as, the channel impulse response from the channel estimation device 7. The matched filter 12 is coupled to a signal power estimator 13 and a channel estimation device 7. Although a matched filter 12 is shown in
The signal power estimator 13, coupled to the matched filter 12 and the comparator 14, receives the output of the matched filter 12 and estimates the signal power of the soft decisions in the received communication. As those skilled in the art know, a method of estimating the signal power is to separate the real and imaginary parts of the outputs of matched filter 12 and calculate the power therefrom. Any method of signal power estimation, though, may be used by the signal power estimator 13. Once the signal power estimator 13 determines the signal power of the soft decisions in the received communication, it is forwarded to the comparator 14.
The comparator 14 is coupled at its inputs to the signal power estimator 13 and the noise power estimator 11, and at its output to the data estimation device 2. The comparator 14 compares the scaled noise power and the signal power and the result of the comparison is used to indicate whether the particular CCTrCh is still in full DTX. For purposes of this disclosure, DTX will be indicative of the full DTX state discussed hereinabove. If the scaled estimated noise power is greater than the estimated signal power for the particular code carrying the TFCI in the first timeslot allocated to the CCTrCh in a frame, the comparator 14 outputs a signal to the data estimation device 2 indicating that no data was sent for the particular CCTrCh. This results is in the data estimation device 2 not operating to demodulate the particular CCTrCh.
If the estimated signal power for the particular code carrying the TFCI in the first timeslot allocated to the CCTrCh in a frame is greater than the scaled estimated noise power, the comparator 14 outputs a signal, to the data estimation device 2 indicating that the end of DTX has been detected, which results in the data estimation device activating the CCTrCh.
In the description above, the comparison between the scaled noise power and the estimated signal power is limited to the particular code carrying the TFCI since if any codes are transmitted then the code carrying the TFCI will be among them. As those skilled in the art know, the comparison can use other received codes allocated to the CCTrCh. If the estimated signal power is greater than the scaled noise power for any particular code, the comparator 14 outputs a signal to the data estimation device 2. The data estimation device 2 can then activate demodulation of the code. Alternatively, it can be activated to demodulate the CCTrCh.
The data estimation device 2, coupled to the demodulator 8, burst detector 10, the channel estimation device 7, and the data demultiplexing and decoding device 4, comprises a code detection device (CDD) 15, a MUD 16, and a TFCI decoder 17. The MUD 16 decodes the received data using the channel impulse responses from the channel estimation device 7 and a set of channelization codes, spreading codes, and channel offsets from the CDD. As those skilled in the art know, the MUD 16 may utilize any multi-user detection method to estimate the data symbols of the received communication, a minimum mean square error block linear equalizer (MMSE-BLE), a zero-forcing block linear equalizer (ZF-BLE) or the use of a plurality of joint detectors, each for detecting one of the plurality of receivable CCTrChs associated with the UE 19.
The CDD 15, coupled to the MUD 16 and the burst detector 10, provides the MUD 16 with the set of codes for each of the plurality of received CCTrChs associated with the receiver 19. If the burst detector 10 indicates that the end of DTX state has been detected, the CDD 15 generates the code information and forwards it to the MUD 16 for decoding of the data. Otherwise, the CDD 15 does nothing with the particular CCTrCh.
Once the MUD 16 has decoded the received data, the data is forwarded to the TFCI decoder 17 and the data demultiplexing and decoding device 4. As those skilled in the art know, the TFCI decoder 17 outputs the maximum-likelihood set of TFCI information bits given the received information. When the value of the TFCI decoder 17 is equal to zero (0), a special burst has been detected, indicating the CCTrCh is beginning DTX or remains in the DTX state.
As stated above, the data estimation device 2 forwards the estimated data to the data demultiplexing and decoding device 4. The demultiplexing and decoding device 4, coupled to the data estimation device 2, detects the received signal to interference ratio (SIR) of the particular CCTrCh or the code carrying the TFCI in the CCTrCh. If the value of the SIR is greater than a predetermined threshold, the end of DTX detected by the burst detector 10 is validated. If the SIR is below the threshold, then a false detection has occurred, indicating that the particular CCTrCh is still in the DTX state. The data demultiplexing and decoding may include error detection on the data which acts as a sanity check for the burst detector 10, reducing the effect of false detections by the UE receiver 19.
The flow diagram of the operation of the receiver in accordance with the preferred embodiment of the present invention is illustrated in
If the burst detector 10 indicates to the CDD 15 that the CCTrCh is in the DTX state, the burst detector 10 continues to monitor the CCTrCh (Step 409). Otherwise, the burst detector indicates to the CDD 15 that the CCTrCh is not in the DTX state (Step 404). The CDD 15 then provides the MUD 16 with the code information for the particular CCTrChs associated with the UE (Step 405). The MUD 16 processes the received CCTrCh and forwards the data symbols to the TFCI decoder 17 and the data demultiplexing and decoding device 4 (Step 406). The TFCI decoder 17 processes the received data symbols to determine the TFCI value (Step 407). If the TFCI value is zero (0), the special burst has been detected and a signal is then sent to the burst detector 10 to continue to monitor the CCTrCh (Step 409), indicating that the CCTrCh is in, or still in, the full DTX state.
If the TFCI value is greater than zero (0), and a CCTrCh is currently in the full DTX state, then the UE performs a sanity check on the received data using information provided by the data demultiplexing and decoding device 4 (Step 408). Referring to
If the sanity check determines that a CCTrCh is in the full DTX state, then an output signal is sent to the burst detector 10 indicating that the burst detector 10 should continue to monitor the CCTrCh to determine when full DTX ends and supply an output to the code detection device 15. If the DTX control logic determines that a CCTrCh is not in the full DTX state then it outputs a signal to the burst detector 10 indicating that it should not monitor the CCTrCh and the decoded data is utilized by the UEs (Step 410).
An alternative embodiment of the burst detector 50 of the present invention is illustrated in
The noise estimator 53, coupled to the TFCI decoder 52, and the comparator 54, receives the decoded TFCI power and the largest TFCI power and calculates a predetermined statistic, such as the root-mean-square of all inputs. The statistic provides an estimate of the noise that the TFCI decoder 52 is subject to. The noise estimate is scaled and forwarded to the comparator 54 for comparison to the largest TFCI power from the TFCI decoder 52.
The comparator 54, coupled to the TFCI decoder 52 and the noise estimator 53, receives the largest TFCI power and the scaled noise estimate and determines the greater of the two values. Similar to the preferred embodiment, if the estimated TFCI power is greater than the scaled noise estimate, the burst detector 50 signals to the data estimation device 2, which activates the CCTrCh demodulation of the particular CCTrCh associated with the UE. Otherwise, the burst detector 50 signals to the data estimation device 2 that the CCTrCh remains in the DTX state.
A second alternative embodiment of the burst detector is illustrated in
The operation of this second alternative is the same as the previous alternative. The matched filter 61 receives the demodulated received signal, determines the soft symbols of the CCTrCh using the first code for the particular CCTrCh and forwards the soft symbols to the TFCI decoder 63. The TFCI decoder 63 decodes the received soft symbols to produce a decoded TFCI word. An estimate of the power of the decoded TFCI word is then generated by the decoder and forwarded to the comparator 64. The comparator 64 receives the power estimate for the decoded TFCI word and a scaled noise estimate from the noise estimator 62 and determines which of the two values is greater. Again, if the estimated power of the TFCI word is greater than the scaled noise estimate, the burst detector 60 signals to the data estimation device 2 that data has been transmitted in the particular CCTrCh associated with the receiver 19, indicative of the end of DTX state or the transmission of the special burst.
A third alternative embodiment of the burst detector is illustrated in
The adder 79, coupled to the symbol power estimator 78, the TFCI decoder 73 and the comparator 74, adds a scaled TFCI power estimate from the TFCI decoder 73 and the scaled symbol power estimate from the symbol power estimator 78, then forwards the summed power estimate to the comparator 74 for comparison to the noise estimate. A determination is then made as to whether data has been transmitted in the CCTrCh. This third alternative embodiment improves the performance of the burst detector 70 with a TFCI detector in those cases where the power estimate of the TFCI word is too low for a reliable determination of the state of the CCTrCh.
A fourth alternative embodiment of the burst detector of the present invention is illustrated in
A fifth alternative embodiment of the burst detector of the present invention is illustrated in
The second matched filter 92, coupled to the demodulator 8 and the comparator 94, receives the demodulated received signal and generates a noise estimate using a ‘nearly’ orthogonal code. The ‘nearly’ orthogonal codes are determined by selecting codes that have low cross correlation with the subset of orthogonal codes used in a particular timeslot where the associated CCTrCh is located. For those systems that do not use all of their orthogonal codes in a timeslot, the ‘nearly’ orthogonal code could be one of the unused orthogonal codes. For example, in a 3GPP TDD or TD-SCDMA system there are 16 OVSF codes. If less than all 16 OVSF codes are used in a timeslot, then the ‘nearly’ orthogonal code would equal one of the unused OVSF codes. The noise estimate generated by the second matched filter 92 is scaled by a predetermined factor and forwarded to the comparator 94.
A sixth alternative embodiment of the burst detector of the present invention is illustrated in
The burst detector of the present invention provides a receiver with the ability to monitor the received signal to determine if a particular CCTrCh associated with the UE has reached the end of full DTX state. In particular, this ability is provided before the data estimation, avoiding the need for the data estimation device to process a large number of codes that may not have been transmitted. This results in a reduction in unnecessary power dissipation during full DTX by not operating the MUD (or other data estimation device) on the particular CCTrCh in the full DTX state. In the case where a CCTrCh is allocated physical channels in multiple timeslots in a frame, and the burst detector has indicated that DTX has not ended, the full receiver chain can remain off during the second and subsequent timeslots in a frame saving significantly more power.
The burst detector also results in better performance by eliminating the occurrence of the filling of the MUD with codes that were not transmitted, which reduces the performance of the CCTrChs associated with the UE. To simplify implementation, code detection devices often assume that at least one code has been transmitted and employ relative power tests to select the set of codes to output to the MUD. If no codes are transmitted for CCTrCh, such as during full DTX, a code detection device may erroneously identify codes as having been transmitted leading to poor performance. By determining whether full DTX is continuing and providing the information to the code detection device, the burst detector allows use of simpler code detection algorithms. Multiple burst detectors can be used in parallel (
While the present invention has been described in terms of the preferred embodiment, other variations which are within the scope of the invention as outlined in the claims below will be apparent to those skilled in the art.
Claims
1. A burst detector for detecting when a selected timeslot which includes data signals for a plurality of channels is received without selected ones of the plurality of channels, the burst detector comprising:
- a comparator for generating a burst detection signal when a signal power estimate of a predetermined code detected within a signal received in a selected timeslot is greater than a scaled noise power estimate of any signal received in the selected timeslot; and
- a data estimation device for decoding the received signal of the timeslot when the burst detection signal is generated.
2. The burst detector of claim 1 wherein the data estimation device comprises:
- a code detection device for generating signal codes in response to the burst detection signal;
- a decoder for decoding a received signal in response to received signal codes; and
- a transport format combination index (TFCI) decoder, coupled to the decoder, for detecting a TFCI signal in a decoded received signal.
3. The burst detector of claim 2 further comprising a demultiplexer responsive to the data estimation device, for verifying that the selected timeslot includes channel data and generating a monitoring signal when channel data is present.
4. The burst detector of claim 3 wherein the burst detector ceases detection of a received signal when a monitoring signal is generated and the TFCI signal indicates that one or more of selected channels have been received in the selected timeslot.
5. The burst detector of claim 4 wherein the burst detector continues to detect the received signal when the TFCI signal indicates that no selected channels have been received in the timeslot.
6. The burst detector of claim 5 wherein the data estimation device comprises:
- a code detection device for generating signal codes in response to the burst detection signal;
- a decoder for decoding a received signal in response to signal codes received from the code detection device; and
- a transport format combination index (TFCI) decoder, coupled to the decoder, for detecting a TFCI signal in a decoded received signal.
7. A code detection method for detecting when a selected timeslot which includes data signals for a plurality of channels is received without selected ones of the plurality of channels, the method comprising:
- generating a burst detection signal when a signal power estimate of a predetermined code detected within a signal received in a selected timeslot is greater than a scaled noise power estimate of any signal received in the selected timeslot; and
- decoding the received signal of the timeslot when the burst detection signal is generated.
8. The method of claim 7 further comprising:
- generating signal codes in responses to the burst detection signal;
- detecting a transport format combination index (TFCI) signal in a received signal representing a number of selected channels in the selected timeslot;
- verifying that the selected timeslot includes channel data; and
- generating a monitoring signal when channel data is present in the selected timeslot.
Type: Application
Filed: Jan 3, 2007
Publication Date: May 17, 2007
Applicant: InterDigital Technology Corporation (Wilmington, DE)
Inventor: Robert DiFazio (Greenlawn, NY)
Application Number: 11/648,931
International Classification: H04B 7/212 (20060101);