US20020010896A1 - Modulation detection method and apparatus for reducing number of potential path metric by selecting out constellation points having smallest error metrics before modulation detector - Google Patents

Modulation detection method and apparatus for reducing number of potential path metric by selecting out constellation points having smallest error metrics before modulation detector Download PDF

Info

Publication number
US20020010896A1
US20020010896A1 US09/439,805 US43980599A US2002010896A1 US 20020010896 A1 US20020010896 A1 US 20020010896A1 US 43980599 A US43980599 A US 43980599A US 2002010896 A1 US2002010896 A1 US 2002010896A1
Authority
US
United States
Prior art keywords
constellation points
path
error metrics
modulation
basis
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
US09/439,805
Other versions
US6353913B2 (en
Inventor
Mikko Huttunen
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
WSOU Investments LLC
Original Assignee
Nokia Networks Oy
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nokia Networks Oy filed Critical Nokia Networks Oy
Assigned to NOKIA NETWORKS OY reassignment NOKIA NETWORKS OY ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: HUTTUNEN, MIKKO
Publication of US20020010896A1 publication Critical patent/US20020010896A1/en
Application granted granted Critical
Publication of US6353913B2 publication Critical patent/US6353913B2/en
Assigned to NOKIA CORPORATION reassignment NOKIA CORPORATION MERGER (SEE DOCUMENT FOR DETAILS). Assignors: NOKIA NETWORKS OY
Assigned to NOKIA TECHNOLOGIES OY reassignment NOKIA TECHNOLOGIES OY ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: NOKIA CORPORATION
Assigned to NOKIA CORPORATION reassignment NOKIA CORPORATION MERGER (SEE DOCUMENT FOR DETAILS). Assignors: NOKIA NETWORKS OY
Assigned to OMEGA CREDIT OPPORTUNITIES MASTER FUND, LP reassignment OMEGA CREDIT OPPORTUNITIES MASTER FUND, LP SECURITY INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: WSOU INVESTMENTS, LLC
Assigned to WSOU INVESTMENTS, LLC reassignment WSOU INVESTMENTS, LLC ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: NOKIA TECHNOLOGIES OY
Anticipated expiration legal-status Critical
Assigned to WSOU INVESTMENTS, LLC reassignment WSOU INVESTMENTS, LLC RELEASE BY SECURED PARTY (SEE DOCUMENT FOR DETAILS). Assignors: OCO OPPORTUNITIES MASTER FUND, L.P. (F/K/A OMEGA CREDIT OPPORTUNITIES MASTER FUND LP
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00Modulated-carrier systems
    • H04L27/0012Modulated-carrier systems arrangements for identifying the type of modulation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00Baseband systems
    • H04L25/02Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006Arrangements for removing intersymbol interference
    • H04L25/03178Arrangements involving sequence estimation techniques
    • H04L25/03203Trellis search techniques
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00Modulated-carrier systems
    • H04L27/18Phase-modulated carrier systems, i.e. using phase-shift keying
    • H04L27/20Modulator circuits; Transmitter circuits
    • H04L27/2032Modulator circuits; Transmitter circuits for discrete phase modulation, e.g. in which the phase of the carrier is modulated in a nominally instantaneous manner
    • H04L27/2053Modulator circuits; Transmitter circuits for discrete phase modulation, e.g. in which the phase of the carrier is modulated in a nominally instantaneous manner using more than one carrier, e.g. carriers with different phases
    • H04L27/206Modulator circuits; Transmitter circuits for discrete phase modulation, e.g. in which the phase of the carrier is modulated in a nominally instantaneous manner using more than one carrier, e.g. carriers with different phases using a pair of orthogonal carriers, e.g. quadrature carriers
    • H04L27/2067Modulator circuits; Transmitter circuits for discrete phase modulation, e.g. in which the phase of the carrier is modulated in a nominally instantaneous manner using more than one carrier, e.g. carriers with different phases using a pair of orthogonal carriers, e.g. quadrature carriers with more than two phase states
    • H04L27/2078Modulator circuits; Transmitter circuits for discrete phase modulation, e.g. in which the phase of the carrier is modulated in a nominally instantaneous manner using more than one carrier, e.g. carriers with different phases using a pair of orthogonal carriers, e.g. quadrature carriers with more than two phase states in which the phase change per symbol period is constrained
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00Baseband systems
    • H04L25/02Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006Arrangements for removing intersymbol interference
    • H04L2025/0335Arrangements for removing intersymbol interference characterised by the type of transmission
    • H04L2025/03375Passband transmission
    • H04L2025/03401PSK

Definitions

  • the invention relates to a modulation detection method comprising taking a sample from a received signal, determining reference constellation points, calculating path-specific error metrics from a sample point to the reference constellation points, and applying the path-specific error metrics corresponding to the reference constellation points to a modulation detector.
  • Modulation converts the signal into a form in which it can be transmitted at radio frequency.
  • a modulation method can be considered efficient for instance if it allows as much information as possible to be transferred at as narrow a frequency band as possible.
  • Modulation should also cause as little interference as possible to adjacent channels.
  • Modulation methods include e.g. ⁇ /4-DQPSK ( ⁇ /4-shifted, Differential Quaternary Phase Shift Keying) modulation.
  • This modulation method comprises eight phase states, but only four phase shifts. Allowed phase shifts (symbols) are ⁇ /4 and ⁇ 3 ⁇ /4.
  • FIG. 3A shows the modulation phase shift diagram (constellation). Each phase shift corresponds to two bits to be transmitted.
  • a digital signal modulates the carrier in two-bit periods so that a given phase shift during each symbol period corresponds to a given two-bit combination.
  • a symbol period refers to a signal period employed in the transmission of two bits.
  • Phase shifts corresponding to bit combinations 00, 01, 10 and 11 are ⁇ /4, 3 ⁇ /4, ⁇ /4 and ⁇ 3 ⁇ /4.
  • the symbol frequency used in e.g. the TETRA system is 18 kHz, the bit frequency being 36 kHz.
  • a signal When a signal is being received, it has to be demodulated in order for the information therein to be detected.
  • a signal transferred over the radio path can be distorted in various ways, thus complicating modulation detection.
  • Signal-impairing phenomena include e.g. noise and inter-symbol interference (ISI).
  • ISI inter-symbol interference
  • a known solution to implement noise-tolerant modulation detection is an MLSE detector (Maximum Likelihood Sequence Estimation), which utilizes the generally known Viterbi algorithm, i.e. sequence detection.
  • the Viterbi algorithm is a method of determining the most probable signal paths on the basis of path error metrics.
  • a signal path refers to different combinations of successive modulation symbols.
  • the MLSE detector such alternative signal paths are created and bits corresponding to a given signal period are detected at a delay based on the error metrics of a signal path composed of a plurality of successive modulation symbols, the path having the best metrics being selected from the alternative signal paths.
  • the interference cancellation provided by the method is the better the more successive symbol periods the paths comprise.
  • detection depth which refers to path length.
  • each symbol period comprises four different symbol alternatives, making the number of paths four-fold within each period. This, in turn, significantly increases the required computational capacity.
  • the objects of the invention are achieved by a method which is characterized by selecting out of the reference constellation points the constellation points, preferably two of them, having the smallest error metrics, and applying only the error metrics corresponding thereto to the modulation detector.
  • the invention is based on selecting, during each symbol period, the constellation points corresponding to symbols closest to a sample point taken from a signal.
  • the sample point can be assumed at a very high probability to be either of the two closest constellation points, and consequently these two points are selected for further processing.
  • the two remaining constellation points are discarded from further processing.
  • An advantage of the method of the invention is that since improbable alternatives are discarded at an initial stage, the calculation required can be reduced. This, in turn, allows an increase in detection depth at a given computational capacity.
  • the invention also relates to an apparatus for modulation detection, the apparatus comprising: means for taking a sample from a signal, means for determining reference constellation points and calculating path-specific error metrics from a sample point to the reference constellation points, the apparatus being adapted to use the path-specific error metrics corresponding to the reference constellation points in the modulation detector for detecting bits, the apparatus being characterized by being adapted to select out of the reference constellation points the constellation points, preferably two of them, having the smallest error metrics and to use only the error metrics corresponding thereto in the modulation detector.
  • Such an apparatus serves to achieve the advantages offered by the modulation detection method of the invention by a simple structure.
  • FIG. 1 is a block diagram of a reception structure according to the TETRA system
  • FIG. 2 is a simplified diagrammatic plan of a frame structure of the TETRA system
  • FIG. 3A is a phase shift diagram of ⁇ /4-DQPSK modulation
  • FIG. 3B shows the constellation points of ⁇ /4-DQPSK modulation
  • FIG. 4 is a block diagram of an adaptive MLSE detector and related channel estimators in accordance with an embodiment
  • FIG. 5 is a trellis diagram of a 16-state detector
  • FIG. 6 is a trellis diagram of a 4-state detector
  • FIG. 7 shows formation of signal paths in the course of four symbol periods with two constellation point alternatives
  • FIG. 8 is a trellis diagram of an embodiment of the invention.
  • information bits received from the management layer MAC are coded by block coding and convolution coding in order to be able to detect and possibly correct errors generated in the signal on the radio path.
  • the coded bits are interleaved in such a way that successive bits are far apart. This facilitates error correction, should the signal to be transmitted be subjected to momentary interference on the radio path.
  • the interleaved bits are mixed by a given colour code allowing the identification of transmissions by different base stations.
  • multiplexing the bits of different logical channels are combined. The multiplexed bits are then used to form a burst.
  • a burst is a structure sent in one TDMA (Time Division Multiple Access) timeslot or subtimeslot.
  • a burst is composed of data bit fields 20 and 22 and a training sequence 21 between them in the middle of the burst as is shown in FIG. 2.
  • Differential coding generates modulating symbols from the bit pairs of a burst.
  • a carrier modulated by symbol control is amplified in a transmitter and transmitted to the radio path.
  • the modulation employed is the above described ⁇ /4-DQPSK ( ⁇ /4-shifted, Differential Quaternary Phase Shift Keying) modulation.
  • This modulation method comprises eight phase states; but only four phase shifts. Allowed phase shifts (symbols) are ⁇ /4 and ⁇ 3 ⁇ /4.
  • the ⁇ /4-DQPSK constellation thus varies at intervals of a symbol between two 4-point constellations, illustrated in FIG. 3B by four black points (1 st constellation) and four white points (2 nd constellation). As the symbol period changes, a shift may occur only from a white point to a black point and from a black point to a white point.
  • Each of these eight constellation points can be represented by numbers 0 to 3, as is shown in FIG. 3B. Non-idealities of a radio channel may cause constellation points to shift.
  • FIG. 1 is a block diagram of a receiver structure according to the invention for e.g. the TETRA system. Only the receiver parts essential to the description of the invention are shown.
  • a signal is received from an antenna (not shown) and radio-frequency parts first process the signal. Samples are then taken with an A/D converter (not shown) from an intermediate-frequency signal. The samples are applied to a synchronization block 11 , as is illustrated in FIG. 1 by a signal RF 1 .
  • the synchronization block 11 searches the obtained samples for the training sequence associated with the frame structure and uses it to accurately determine the sampling moment, i.e. locations of all symbols in the sample flow.
  • the synchronization block also controls the radio-frequency parts of the receiver so as to maintain a signal arriving at the A/ 0 D converter at an optimal level.
  • the synchronization block applies the frame to a channel equalizer and detector block 14 .
  • the channel equalizer equalizes non-idealities caused by the radio path channel and the associated detector detects the information bits.
  • a logical channel is formed from the frame in framing 18 and sent to further processing.
  • the MLSE detector is provided with the Viterbi algorithm.
  • the Viterbi detector 41 estimates a transmitted sequence r n by means of a description created by the channel estimator 42 a of the impulse response of the radio channel.
  • the channel estimator 42 a adaptively estimates the impulse response of the radio channel by means of decisions J n produced by the Viterbi detector 41 or by means of preliminary decisions.
  • one channel estimate corresponds to each sequence of the Viterbi detector. These estimates can be realized by one common channel estimator, but this would impair the tracing capability of the channel estimator.
  • the embodiment shown in FIG. 4 comprises a plurality of parallel channel estimators 42 a , preferably as many as there are sequences.
  • a transmitted signal arrives at a receiver along a plurality of propagation paths, each having a specific time delay, channel properties also changing as a function of time.
  • beams reflected and delayed on the radio path cause so-called inter-symbol interference (ISI).
  • ISI inter-symbol interference
  • the frequency response or impulse response of a channel can be estimated by a discrete-timed filter, a channel estimator, whose tap coefficients model the radio channel.
  • a channel estimator is used to describe the state of a radio channel.
  • a channel estimator refers generally to a mechanism for estimating and maintaining a description of the complex impulse response of a radio channel.
  • a method of updating the channel estimate is an essential part of this mechanism.
  • the LMS (Least Mean Square) algorithm can be used to update channel estimates.
  • the detector 14 In order to ensure the convergence of the LMS algorithm before the beginning of actual information bits, the detector 14 has to get an optimal initial estimate for the state of the channel. This estimate is obtained from the synchronization 11 , which, when searching for an optimal sampling moment, calculates a complex cross correlation between the training sequence 21 of a received signal and a stored version of the training sequence.
  • the Viterbi algorithm is a method of searching for a trellis for a signal path corresponding to maximum probability among a number of signal paths, in which one channel estimate corresponds to each signal path.
  • M L sequences propagate in the detector, each of which has specific path metrics based on Euclidean distance.
  • the main idea of MLSE detection is the attempt to construct a reference constellation point on the basis of information on, i.e. the channel estimate for, the current state of the channel. Once the reference constellation point has been calculated and selected on the basis of the best signal path, the difference between the reference point and the received sample can be calculated for each channel estimate. This error can be used to update the channel estimator.
  • FIGS. 5 and 6 clearly show how the length or depth L of the detection significantly affects the complexity of the calculation. On the other hand, the longer the length L, the better the accuracy in selection of the best path.
  • the calculation required can be reduced by selecting from the reference constellation points defined at a given stage of the trellis search a few, preferably two, constellation points having the least error metrics. At the next stage of the trellis search, only these selected reference points will be used instead of all of them, as in conventional MLSE detectors. Less calculation can be directly utilized as lower performance requirements for the equipment (usually a signal processor), lowering the price and power consumption of the equipment, or as calculation of longer signal paths, improving the accuracy of the detection.
  • a synchronization block applies an initial estimate for the radio channel to an estimator.
  • Backward detection over a training sequence 21 and a first half timeslot 20 and, correspondingly, forward detection over the training sequence 21 and a second half timeslot 22 is then performed.
  • the MLSE detector starts to detect the bits of the training sequence 21 backwards starting from the end of the training sequence towards the beginning of the burst, as FIG. 2 illustrates.
  • the MLSE detector forms alternative signal paths, the number depending on the detection depth employed.
  • reference constellation points are determined on the basis of the channel estimate corresponding to the best known signal path.
  • the best known path is the path whose error metrics (the combined error metrics of the points on the path) are the lowest, and accordingly the corresponding channel estimate. is probably the closest to the actual state of the channel.
  • error metrics from the sample point to each of the four reference constellation points are calculated.
  • two points having the least error metrics are selected for further study.
  • the constellation point having the best metrics N over a symbol period is denoted by the numeral 0 and the second best constellation by the numeral 1.
  • Path-specific reference constellation points corresponding to these two selected points are determined for the other paths. This reduces the need for calculation, since all four reference constellation points have to be determined only for the best path, and only two for the other paths.
  • the error metrics of the paths are calculated by adding the error metrics corresponding to the two selected constellation points, each separately, to the error metrics of the paths.
  • FIG. 7 shows a diagram of the formation of signal paths in a situation where the method of the invention is being used.
  • N is the number of symbol periods to be taken into consideration.
  • 16 different path alternatives are obtained during four symbol periods, these being: 1: 0000 5: 0100 9: 1000 13: 1100 2: 0001 6: 0101 10: 1001 14: 1101 3: 0010 7: 0110 11: 1010 15: 1110 4: 0011 8: 0111 12: 1011 16: 1111
  • FIG. 8 shows a trellis diagram illustrating system states and state shifts in the above described case where the method of the invention is used in accordance with an embodiment.
  • the MLSE detector starts to detect the bits of the training sequence 21 forward starting from the beginning of the training sequence towards the end of the burst, as FIG. 2 illustrates.
  • the search for the best path continues as described above until the end of the second half timeslot 22 is reached.
  • the path having the best metrics over the entire second half timeslot 22 of the burst is selected, and this path is then traced back up to the end of the training sequence 21 .
  • Symbol detection is carried out during the backtracing procedure.

Abstract

A modulation detection method and apparatus, comprising taking a sample from a received signal, determining reference constellation points, calculating path-specific error metrics from a sample point to the reference constellation points, and applying the path-specific error metrics corresponding to the reference constellation points to a modulation detector. Out of the reference constellation points the constellation points, preferably two of them, having the smallest error metrics are selected, and only the error metrics corresponding thereto are applied to the modulation detector.

Description

    BACKGROUND OF THE INVENTION
  • The invention relates to a modulation detection method comprising taking a sample from a received signal, determining reference constellation points, calculating path-specific error metrics from a sample point to the reference constellation points, and applying the path-specific error metrics corresponding to the reference constellation points to a modulation detector. [0001]
  • When information is transferred on a radio channel, the signal to be transmitted has to be subjected to modulation. Modulation converts the signal into a form in which it can be transmitted at radio frequency. A modulation method can be considered efficient for instance if it allows as much information as possible to be transferred at as narrow a frequency band as possible. Depending on the purpose of use, other features can also be emphasized. Modulation should also cause as little interference as possible to adjacent channels. [0002]
  • Modulation methods include e.g. π/4-DQPSK (π/4-shifted, Differential Quaternary Phase Shift Keying) modulation. This modulation method comprises eight phase states, but only four phase shifts. Allowed phase shifts (symbols) are ±π/4 and ±3π/4. FIG. 3A shows the modulation phase shift diagram (constellation). Each phase shift corresponds to two bits to be transmitted. In other words, a digital signal modulates the carrier in two-bit periods so that a given phase shift during each symbol period corresponds to a given two-bit combination. A symbol period refers to a signal period employed in the transmission of two bits. Phase shifts corresponding to [0003] bit combinations 00, 01, 10 and 11 are π/4, 3π/4, −π/4 and −3π/4. The symbol frequency used in e.g. the TETRA system (Terrestrial Trunked Radio) is 18 kHz, the bit frequency being 36 kHz.
  • When a signal is being received, it has to be demodulated in order for the information therein to be detected. However, a signal transferred over the radio path can be distorted in various ways, thus complicating modulation detection. Signal-impairing phenomena include e.g. noise and inter-symbol interference (ISI). A known solution to implement noise-tolerant modulation detection is an MLSE detector (Maximum Likelihood Sequence Estimation), which utilizes the generally known Viterbi algorithm, i.e. sequence detection. The Viterbi algorithm is a method of determining the most probable signal paths on the basis of path error metrics. Herein a signal path refers to different combinations of successive modulation symbols. In the MLSE detector such alternative signal paths are created and bits corresponding to a given signal period are detected at a delay based on the error metrics of a signal path composed of a plurality of successive modulation symbols, the path having the best metrics being selected from the alternative signal paths. The interference cancellation provided by the method is the better the more successive symbol periods the paths comprise. In fact, in this connection the term employed is detection depth, which refers to path length. [0004]
  • The problem in the above solution is that when the detection depth, i.e. the number of symbol periods included in the paths, is increased, the number of paths also increases. When e.g. π/4-DQPSK modulation is used, each symbol period comprises four different symbol alternatives, making the number of paths four-fold within each period. This, in turn, significantly increases the required computational capacity. [0005]
  • BRIEF DESCRIPTION OF THE INVENTION
  • It is an object of the invention to provide a method which solves the above problems. The objects of the invention are achieved by a method which is characterized by selecting out of the reference constellation points the constellation points, preferably two of them, having the smallest error metrics, and applying only the error metrics corresponding thereto to the modulation detector. [0006]
  • The invention is based on selecting, during each symbol period, the constellation points corresponding to symbols closest to a sample point taken from a signal. In e.g. π/4-DOPSK modulation, the sample point can be assumed at a very high probability to be either of the two closest constellation points, and consequently these two points are selected for further processing. The two remaining constellation points are discarded from further processing. [0007]
  • An advantage of the method of the invention is that since improbable alternatives are discarded at an initial stage, the calculation required can be reduced. This, in turn, allows an increase in detection depth at a given computational capacity. [0008]
  • The invention also relates to an apparatus for modulation detection, the apparatus comprising: means for taking a sample from a signal, means for determining reference constellation points and calculating path-specific error metrics from a sample point to the reference constellation points, the apparatus being adapted to use the path-specific error metrics corresponding to the reference constellation points in the modulation detector for detecting bits, the apparatus being characterized by being adapted to select out of the reference constellation points the constellation points, preferably two of them, having the smallest error metrics and to use only the error metrics corresponding thereto in the modulation detector. Such an apparatus serves to achieve the advantages offered by the modulation detection method of the invention by a simple structure.[0009]
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • The invention will be described. in greater detail by means of preferred embodiments with reference to the attached drawings, in which: [0010]
  • FIG. 1 is a block diagram of a reception structure according to the TETRA system; [0011]
  • FIG. 2 is a simplified diagrammatic plan of a frame structure of the TETRA system; [0012]
  • FIG. 3A is a phase shift diagram of π/4-DQPSK modulation; [0013]
  • FIG. 3B shows the constellation points of π/4-DQPSK modulation; [0014]
  • FIG. 4 is a block diagram of an adaptive MLSE detector and related channel estimators in accordance with an embodiment; [0015]
  • FIG. 5 is a trellis diagram of a 16-state detector; [0016]
  • FIG. 6 is a trellis diagram of a 4-state detector, [0017]
  • FIG. 7 shows formation of signal paths in the course of four symbol periods with two constellation point alternatives; [0018]
  • FIG. 8 is a trellis diagram of an embodiment of the invention.[0019]
  • DETAILED DESCRIPTION OF THE INVENTION
  • In the following the invention will be described in the TETRA system, but the invention is not to be restricted to any given system or modulation method. [0020]
  • In the TETRA system, information bits received from the management layer MAC (Medium Access Layer) are coded by block coding and convolution coding in order to be able to detect and possibly correct errors generated in the signal on the radio path. The coded bits are interleaved in such a way that successive bits are far apart. This facilitates error correction, should the signal to be transmitted be subjected to momentary interference on the radio path. The interleaved bits are mixed by a given colour code allowing the identification of transmissions by different base stations. In multiplexing, the bits of different logical channels are combined. The multiplexed bits are then used to form a burst. A burst is a structure sent in one TDMA (Time Division Multiple Access) timeslot or subtimeslot. A burst is composed of [0021] data bit fields 20 and 22 and a training sequence 21 between them in the middle of the burst as is shown in FIG. 2. Differential coding generates modulating symbols from the bit pairs of a burst. A carrier modulated by symbol control is amplified in a transmitter and transmitted to the radio path.
  • The modulation employed is the above described π/4-DQPSK (π/4-shifted, Differential Quaternary Phase Shift Keying) modulation. This modulation method comprises eight phase states; but only four phase shifts. Allowed phase shifts (symbols) are ±π/4 and ±3π/4. In practice the π/4-DQPSK constellation thus varies at intervals of a symbol between two 4-point constellations, illustrated in FIG. 3B by four black points (1[0022] st constellation) and four white points (2nd constellation). As the symbol period changes, a shift may occur only from a white point to a black point and from a black point to a white point. Each of these eight constellation points can be represented by numbers 0 to 3, as is shown in FIG. 3B. Non-idealities of a radio channel may cause constellation points to shift.
  • FIG. 1 is a block diagram of a receiver structure according to the invention for e.g. the TETRA system. Only the receiver parts essential to the description of the invention are shown. Upon reception, a signal is received from an antenna (not shown) and radio-frequency parts first process the signal. Samples are then taken with an A/D converter (not shown) from an intermediate-frequency signal. The samples are applied to a [0023] synchronization block 11, as is illustrated in FIG. 1 by a signal RF1. The synchronization block 11 searches the obtained samples for the training sequence associated with the frame structure and uses it to accurately determine the sampling moment, i.e. locations of all symbols in the sample flow. The synchronization block also controls the radio-frequency parts of the receiver so as to maintain a signal arriving at the A/0D converter at an optimal level. The synchronization block applies the frame to a channel equalizer and detector block 14. The channel equalizer equalizes non-idealities caused by the radio path channel and the associated detector detects the information bits. Finally a logical channel is formed from the frame in framing 18 and sent to further processing.
  • In the above an example of the general structure of a receiver has been described to facilitate understanding the invention. However, the structure of the receiver may change without deviating from the present invention, which is directed to the MLSE detector of a receiver. [0024]
  • In a primary embodiment of the invention the MLSE detector is provided with the Viterbi algorithm. An adaptive MLSE detector thus comprises a [0025] Viterbi detector 41 and at least one adaptive channel estimator 42 a (a=1 to MV), as is shown in FIG. 4. The Viterbi detector 41 estimates a transmitted sequence rn by means of a description created by the channel estimator 42 a of the impulse response of the radio channel. The channel estimator 42 a adaptively estimates the impulse response of the radio channel by means of decisions Jn produced by the Viterbi detector 41 or by means of preliminary decisions. In accordance with the invention, one channel estimate corresponds to each sequence of the Viterbi detector. These estimates can be realized by one common channel estimator, but this would impair the tracing capability of the channel estimator. The embodiment shown in FIG. 4 comprises a plurality of parallel channel estimators 42 a, preferably as many as there are sequences.
  • It is typical of the radio path that a transmitted signal arrives at a receiver along a plurality of propagation paths, each having a specific time delay, channel properties also changing as a function of time. E.g. beams reflected and delayed on the radio path cause so-called inter-symbol interference (ISI). The frequency response or impulse response of a channel can be estimated by a discrete-timed filter, a channel estimator, whose tap coefficients model the radio channel. A channel estimator is used to describe the state of a radio channel. [0026]
  • In the present description a channel estimator refers generally to a mechanism for estimating and maintaining a description of the complex impulse response of a radio channel. A method of updating the channel estimate is an essential part of this mechanism. In the TETRA system, the LMS (Least Mean Square) algorithm can be used to update channel estimates. In order to ensure the convergence of the LMS algorithm before the beginning of actual information bits, the [0027] detector 14 has to get an optimal initial estimate for the state of the channel. This estimate is obtained from the synchronization 11, which, when searching for an optimal sampling moment, calculates a complex cross correlation between the training sequence 21 of a received signal and a stored version of the training sequence. As a result of the cross correlation result, an initial value is obtained for the channel estimate, the value depicting the average state of the channel during a training sequence. Channel equalization and symbol detection is never started until the training sequence has been received. This is because symbol synchronization is able to adjust symbol timing as accurately as possible and generate an initial channel estimate. Both forward and backward channel equalization is performed after estimate initialization by always training the detector 14 across the training sequence 21 towards the end of the burst or, correspondingly, towards the beginning, as FIG. 2 shows.
  • The Viterbi algorithm is a method of searching for a trellis for a signal path corresponding to maximum probability among a number of signal paths, in which one channel estimate corresponds to each signal path. At each stage of the trellis search, M[0028] L sequences propagate in the detector, each of which has specific path metrics based on Euclidean distance. The main idea of MLSE detection is the attempt to construct a reference constellation point on the basis of information on, i.e. the channel estimate for, the current state of the channel. Once the reference constellation point has been calculated and selected on the basis of the best signal path, the difference between the reference point and the received sample can be calculated for each channel estimate. This error can be used to update the channel estimator. Because computational complexity increases exponentially as a function of the length L, a practical implementation of the detector is only feasible at relatively low values of M and L. In e.g. cellular radio systems in which speech is transferred, low values for M and L are often selected, e.g. M=4 and L=2.
  • In the constellation of FIG. 3B, for example, M=4. If the channel estimate depicts the combined effect of a previous, current and coming symbol, then L=2. In this case the number of different constellation point combinations, i.e. states, is 16. This requires a 16-state detector whose trellis diagram is shown in FIG. 5. Shifts are shown from the white constellation to the black one. Consequently, each state will include four paths, the best of which will be selected for further processing. [0029]
  • FIG. 6 shows the trellis diagram of a 4-state Viterbi detector, in which the constellation varies at intervals of a symbol between two four-point constellations, as is the case in a 16-state detector, too (M=4). The definition of a state differs from a 16-state detector in that the state is determined only by the current constellation point, not two successive ones, i.e. L=1. Black points depict alternative modulation symbols and are numbered from 0 to 3. FIG. 6 shows that when the method of the invention is not used, 4*4=16 paths form during only two symbol periods. [0030] Paths alternatives 1 to 16 are:
    1: 00 5: 01  9: 02 13: 03
    2: 10 6: 11 10: 12 14: 13
    3: 20 7: 21 11: 22 15: 23
    4: 30 8: 31 12: 32 16: 33
  • FIGS. 5 and 6 clearly show how the length or depth L of the detection significantly affects the complexity of the calculation. On the other hand, the longer the length L, the better the accuracy in selection of the best path. [0031]
  • In accordance with the invention, the calculation required can be reduced by selecting from the reference constellation points defined at a given stage of the trellis search a few, preferably two, constellation points having the least error metrics. At the next stage of the trellis search, only these selected reference points will be used instead of all of them, as in conventional MLSE detectors. Less calculation can be directly utilized as lower performance requirements for the equipment (usually a signal processor), lowering the price and power consumption of the equipment, or as calculation of longer signal paths, improving the accuracy of the detection. [0032]
  • In the following the detection of the invention will be described by means of an example. [0033]
  • At first a synchronization block applies an initial estimate for the radio channel to an estimator. Backward detection over a [0034] training sequence 21 and a first half timeslot 20 and, correspondingly, forward detection over the training sequence 21 and a second half timeslot 22 is then performed.
  • In backward detection, the MLSE detector starts to detect the bits of the [0035] training sequence 21 backwards starting from the end of the training sequence towards the beginning of the burst, as FIG. 2 illustrates. On the basis of the training sequence the MLSE detector forms alternative signal paths, the number depending on the detection depth employed. When the first IQ sample has been taken after the training sequence when detecting actual information bits, reference constellation points are determined on the basis of the channel estimate corresponding to the best known signal path. The best known path is the path whose error metrics (the combined error metrics of the points on the path) are the lowest, and accordingly the corresponding channel estimate. is probably the closest to the actual state of the channel.
  • Next, error metrics from the sample point to each of the four reference constellation points are calculated. Of these four reference constellation points, two points having the least error metrics are selected for further study. When the best two constellation points have been selected on the basis of the channel estimate for the best path, the constellation point having the best metrics N over a symbol period is denoted by the [0036] numeral 0 and the second best constellation by the numeral 1. Path-specific reference constellation points corresponding to these two selected points are determined for the other paths. This reduces the need for calculation, since all four reference constellation points have to be determined only for the best path, and only two for the other paths. Next the error metrics of the paths are calculated by adding the error metrics corresponding to the two selected constellation points, each separately, to the error metrics of the paths. Consequently, two new paths are obtained for each old path. Finally the obtained path-specific error metrics are applied to a Viterbi detector, which selects the best paths for further processing. Thus the number of active paths remains constant, since the detector discards half of the paths. The above described process continues starting with the taking of a new sample IQ until all information bits of the first half timeslot have been gone through.
  • FIG. 7 shows a diagram of the formation of signal paths in a situation where the method of the invention is being used. There are only two modulation symbols, labelled 0 and 1, for each symbol period, the number of paths being 2[0037] N, wherein N is the number of symbol periods to be taken into consideration. In the case of FIG. 7, 16 different path alternatives are obtained during four symbol periods, these being:
    1: 0000 5: 0100  9: 1000 13: 1100
    2: 0001 6: 0101 10: 1001 14: 1101
    3: 0010 7: 0110 11: 1010 15: 1110
    4: 0011 8: 0111 12: 1011 16: 1111
  • FIG. 8 shows a trellis diagram illustrating system states and state shifts in the above described case where the method of the invention is used in accordance with an embodiment. [0038]
  • On the basis of the above it can be stated that when the required computational capacity depends on the number of paths to be processed, detection depth can be increased significantly at a given computational capacity by discarding improbable symbol alternatives at the start of processing. [0039]
  • When the above described process has reached the beginning of the [0040] first half timeslot 20, the path having the best metrics over the entire first half timeslot 20 of the burst is selected out of the active paths, and this path is then traced back up to the beginning of the training sequence 21. Symbol detection is carried out during the backtracing procedure.
  • In forward detection, the MLSE detector starts to detect the bits of the [0041] training sequence 21 forward starting from the beginning of the training sequence towards the end of the burst, as FIG. 2 illustrates. The search for the best path continues as described above until the end of the second half timeslot 22 is reached. Then the path having the best metrics over the entire second half timeslot 22 of the burst is selected, and this path is then traced back up to the end of the training sequence 21. Symbol detection is carried out during the backtracing procedure.
  • At this point all bits of the burst have been detected. [0042]
  • Even though this embodiment description describes the utilization of the method of the invention in the TETRA system employing π/4-DQPSK modulation, it does in no way restrict the use of the method of the invention also in other types of systems. Consequently, the modulation method employed can be another method and similarly, another detection algorithm may be used instead of the Viterbi algorithm. [0043]
  • It is obvious to a person skilled in the art that as technology advances, the inventive concept can be implemented in several different ways. Thus the invention and its embodiments are not limited to the examples described above, but may vary within the scope of the claims. [0044]

Claims (9)

1. A modulation detection method comprising taking a sample from a received signal, determining reference constellation points, calculating path-specific error metrics from a sample point to the reference constellation points, and applying the path-specific error metrics corresponding to the reference constellation points to a modulation detector, characterized by selecting out of the reference constellation points the constellation points, preferably two of them, having the smallest error metrics, and applying only the error metrics corresponding thereto to the modulation detector.
2. A method as claimed in claim 1, characterized by selecting the constellation points, the error-metrics corresponding to which are applied to the modulation detector, for all paths on the basis of the error metrics corresponding to the reference constellation points determined on the basis of a channel estimate corresponding to the best known path.
3. A method as claimed in claim 2, characterized by determining the reference constellation points path-specifically on the basis of the channel estimate corresponding to each path.
4. A method as claimed in claim 3, characterized by determining for the other paths only path-specific reference constellation points which correspond to the selected constellation points and the corresponding error metrics, after the constellation points having the smallest error metrics have been selected on the basis of error metrics corresponding to the reference constellation points determined on the basis of the channel estimate corresponding to the best known path.
5. A method as claimed in any one of claims 1 to 4, characterized by the modulation detector being a Viterbi detector.
6. A method as claimed in any one of claims 1 to 5, characterized by reading from a received signal a training sequence (21), forming initial estimates for the state of the channel on the basis of the training sequence, reading a first half timeslot (20), forming alternative signal paths, selecting the signal path having the best error metrics, detecting the bits corresponding to the best signal path in the first half timeslot, reading from the received signal the training sequence (21), forming initial estimates for the state of the channel on the basis of the training sequence, reading a second half timeslot (22), forming alternative signal paths, selecting the signal path having the best error metrics, and detecting the bits corresponding the best signal path in the second half timeslot.
7. An apparatus for modulation detection, the apparatus comprising: means for taking a sample from a signal (RF1), means (14) for determining reference constellation points and calculating path-specific error metrics from a sample point to the reference constellation points, the apparatus being adapted to use the path-specific error metrics corresponding to the reference constellation points in the modulation detector (14) for bit detection, characterized by being adapted to select out of the reference constellation points the constellation points, preferably two of them, having the smallest error metrics and to use only the error metrics corresponding thereto in the modulation detector (14).
8. An apparatus as claimed in claim 7, characterized by being adapted to select the constellation points, the error metrics corresponding to which being applied to the modulation detector, for all paths on the basis of the error metrics corresponding to the reference constellation points determined on the basis of a channel estimate corresponding to the best known path.
9. An apparatus as claimed in claim 8, characterized by being adapted to determine the reference constellation points path-specifically on the basis of a channel estimate corresponding to each path.
US09/439,805 1998-03-23 1999-11-12 Modulation detection method and apparatus Expired - Lifetime US6353913B2 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
FI980652 1998-03-23
FI980652A FI104774B (en) 1998-03-23 1998-03-23 Method and apparatus for modulation detection
PCT/FI1999/000218 WO1999049634A2 (en) 1998-03-23 1999-03-22 Modulation detection method and apparatus

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
PCT/FI1999/000218 Continuation WO1999049634A2 (en) 1998-03-23 1999-03-22 Modulation detection method and apparatus

Publications (2)

Publication Number Publication Date
US20020010896A1 true US20020010896A1 (en) 2002-01-24
US6353913B2 US6353913B2 (en) 2002-03-05

Family

ID=8551355

Family Applications (1)

Application Number Title Priority Date Filing Date
US09/439,805 Expired - Lifetime US6353913B2 (en) 1998-03-23 1999-11-12 Modulation detection method and apparatus

Country Status (9)

Country Link
US (1) US6353913B2 (en)
EP (1) EP0983670A2 (en)
CN (1) CN1307774A (en)
AU (1) AU2839399A (en)
CA (1) CA2289754A1 (en)
FI (1) FI104774B (en)
NO (1) NO995710L (en)
NZ (1) NZ501101A (en)
WO (1) WO1999049634A2 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020094041A1 (en) * 2001-01-15 2002-07-18 Robert John Kopmeiners Maximum likelihood detection method using a sequence estimation receiver
US20030185320A1 (en) * 2002-03-25 2003-10-02 Nobufumi Ueno Diversity circuit demodulating OFDM-method signals and diversity receiving apparatus having said diversity circuit therein
US20060291590A1 (en) * 2005-06-23 2006-12-28 Imtinan Elahi Methods and apparatus to compensate for I/Q mismatch in quadrature receivers
US20080063098A1 (en) * 2006-09-13 2008-03-13 Sunplus Technology Co., Ltd. Channel estimation apparatus with the optimal search and method thereof
WO2015093841A1 (en) * 2013-12-19 2015-06-25 한국해양과학기술원 Device and method for determining channel equalization and reception data of receiver for packet communication

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FI112739B (en) 1998-05-25 2003-12-31 Nokia Corp Method and apparatus for detecting an interfering signal in a radio receiver
US6674815B2 (en) * 1999-06-16 2004-01-06 Ericsson, Inc Method for symbol-spaced estimation and/or tracking of a fractionally-spaced fading radio channel
MXPA04008840A (en) * 2002-03-19 2004-11-26 Thomson Licensing Sa Slicing algorithm for multi-level modulation equalizing schemes.
DE102007053091A1 (en) 2007-11-07 2009-05-14 Rohde & Schwarz Gmbh & Co. Kg Method and device for decoding convolutionally coded signals
CN101599772B (en) * 2008-06-04 2013-04-10 联发科技股份有限公司 Method and device for demodulating signal
CN103297362B (en) * 2013-06-05 2016-08-31 上海高清数字科技产业有限公司 A kind of channel estimation system based on error update and method

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4247944A (en) * 1978-11-15 1981-01-27 Ricoh Co., Ltd. V.29 Constellation detection method and apparatus
US4884272A (en) * 1988-02-10 1989-11-28 Mcconnell Peter R H Maximum likelihood diversity receiver
US4937841A (en) * 1988-06-29 1990-06-26 Bell Communications Research, Inc. Method and circuitry for carrier recovery for time division multiple access radio systems
US5594742A (en) * 1990-12-20 1997-01-14 Communications Satellite Corporation Bidirectional trellis coding
US5384810A (en) * 1992-02-05 1995-01-24 At&T Bell Laboratories Modulo decoder
US5315617A (en) * 1992-05-29 1994-05-24 General Electric Company QAM encoding for high-definition television system
US5588028A (en) * 1993-02-02 1996-12-24 U.S. Robotics Simplified trellis decoder
JP2792812B2 (en) 1993-07-02 1998-09-03 沖電気工業株式会社 Maximum likelihood sequence estimator
US5488635A (en) * 1993-10-29 1996-01-30 General Electric Company Low complexity adaptive equalizer radio receiver employing reduced complexity branch metric calculation
US5467374A (en) * 1993-10-29 1995-11-14 General Electric Company Low complexity adaptive equalizer for U.S. digital cellular radio receivers
US5592514A (en) 1995-03-08 1997-01-07 Lucent Technologies Inc. Method of performing signal reconstruction at the receiving end of a communications system, such as for GSM
FR2736231A1 (en) * 1995-06-28 1997-01-03 Trt Telecom Radio Electr DIGITAL COMMUNICATION SYSTEM COMPRISING A RECEIVER HAVING A RHYTHM RECOVERY DEVICE
GB2305827B (en) * 1995-09-30 2000-01-19 Motorola Ltd Viterbi co-processor and method of operation therefor
EP0786872A2 (en) 1995-12-29 1997-07-30 Lucent Technologies Inc. Viterbi decoder with reduced metric computation
US6141384A (en) * 1997-02-14 2000-10-31 Philips Electronics North America Corporation Decoder for trellis encoded interleaved data stream and HDTV receiver including such a decoder
US6005898A (en) * 1997-03-12 1999-12-21 Interdigital Technology Corporation Multichannel viterbi decoder

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020094041A1 (en) * 2001-01-15 2002-07-18 Robert John Kopmeiners Maximum likelihood detection method using a sequence estimation receiver
US7319725B2 (en) * 2001-01-15 2008-01-15 Agere Systems Inc. Maximum likelihood detection method using a sequence estimation receiver
US8665998B2 (en) 2001-01-15 2014-03-04 Agere Systems Llc Maximum likelihood detection method using a sequence estimation receiver
US20030185320A1 (en) * 2002-03-25 2003-10-02 Nobufumi Ueno Diversity circuit demodulating OFDM-method signals and diversity receiving apparatus having said diversity circuit therein
US7369622B2 (en) * 2002-03-25 2008-05-06 Sanyo Electric Co., Ltd. Diversity circuit demodulating OFDM-method signals and diversity receiving apparatus having said diversity circuit therein
US20060291590A1 (en) * 2005-06-23 2006-12-28 Imtinan Elahi Methods and apparatus to compensate for I/Q mismatch in quadrature receivers
US7876856B2 (en) * 2005-06-23 2011-01-25 Texas Instrumentals Incorporated Quadrature receiver with correction engine, coefficient controller and adaptation engine
US20080063098A1 (en) * 2006-09-13 2008-03-13 Sunplus Technology Co., Ltd. Channel estimation apparatus with the optimal search and method thereof
US7936844B2 (en) * 2006-09-13 2011-05-03 Sunplus Technology Co., Ltd. Channel estimation apparatus with the optimal search and method thereof
WO2015093841A1 (en) * 2013-12-19 2015-06-25 한국해양과학기술원 Device and method for determining channel equalization and reception data of receiver for packet communication
KR101741512B1 (en) * 2013-12-19 2017-05-31 한국해양과학기술원 Device and method for equalizing channel and deciding received data in receiver for packet data communication
US9954697B2 (en) 2013-12-19 2018-04-24 Korea Institute Of Ocean Science Technology Device and method for determining channel equalization and reception data of receiver for packet communication

Also Published As

Publication number Publication date
NO995710L (en) 2000-01-21
FI980652A0 (en) 1998-03-23
CA2289754A1 (en) 1999-09-30
NZ501101A (en) 2001-01-26
WO1999049634A3 (en) 1999-11-11
NO995710D0 (en) 1999-11-22
WO1999049634A2 (en) 1999-09-30
FI104774B (en) 2000-03-31
US6353913B2 (en) 2002-03-05
CN1307774A (en) 2001-08-08
FI980652A (en) 1999-09-24
EP0983670A2 (en) 2000-03-08
AU2839399A (en) 1999-10-18

Similar Documents

Publication Publication Date Title
US5621764A (en) Soft decision signal outputting receiver
US5862192A (en) Methods and apparatus for equalization and decoding of digital communications channels using antenna diversity
US5371471A (en) Low complexity adaptive equalizer radio receiver employing direct reference state updates
US5774504A (en) Equalization and decoding for digital communication channel
US5488635A (en) Low complexity adaptive equalizer radio receiver employing reduced complexity branch metric calculation
KR970007362B1 (en) Equalizing device in receiver
US6449320B1 (en) Equalization with DC-offset compensation
US8259862B2 (en) Receivers and symbol decoders thereof
US5400362A (en) Double sided slot traversing decoding for time division multiple access (TDMA) radio systems
US6154507A (en) System and method for signal demodulation
US5235621A (en) Receiver systems
US5467374A (en) Low complexity adaptive equalizer for U.S. digital cellular radio receivers
CA2274101A1 (en) Method and apparatus for bidirectional demodulation of digitally modulated signals
US5550868A (en) π/4-DQPSK delay spread detection and compensation apparatus and method
US6353913B2 (en) Modulation detection method and apparatus
EP2131543B1 (en) Receiver architecture
US6314148B1 (en) Synchronization tracking method
US6438185B1 (en) Self-optimizing channel equalization and detection
KR100626103B1 (en) Receiver for a digital transmission system
EP1119919B1 (en) Method for selecting modulation detector in receiver, and receiver
US7016434B1 (en) Detection of interfering signal in radio receiver
JP3329002B2 (en) Diversity receiver
FI105505B (en) Method and arrangement for expression of modulation
JPH1051363A (en) Viterbi equalizer

Legal Events

Date Code Title Description
AS Assignment

Owner name: NOKIA NETWORKS OY, FINLAND

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HUTTUNEN, MIKKO;REEL/FRAME:010392/0796

Effective date: 19991102

STCF Information on status: patent grant

Free format text: PATENTED CASE

FPAY Fee payment

Year of fee payment: 4

FPAY Fee payment

Year of fee payment: 8

FPAY Fee payment

Year of fee payment: 12

AS Assignment

Owner name: NOKIA CORPORATION, FINLAND

Free format text: MERGER;ASSIGNOR:NOKIA NETWORKS OY;REEL/FRAME:035789/0917

Effective date: 20070911

AS Assignment

Owner name: NOKIA TECHNOLOGIES OY, FINLAND

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:NOKIA CORPORATION;REEL/FRAME:035796/0819

Effective date: 20150116

Owner name: NOKIA CORPORATION, FINLAND

Free format text: MERGER;ASSIGNOR:NOKIA NETWORKS OY;REEL/FRAME:035842/0359

Effective date: 20070911

AS Assignment

Owner name: OMEGA CREDIT OPPORTUNITIES MASTER FUND, LP, NEW YORK

Free format text: SECURITY INTEREST;ASSIGNOR:WSOU INVESTMENTS, LLC;REEL/FRAME:043966/0574

Effective date: 20170822

Owner name: OMEGA CREDIT OPPORTUNITIES MASTER FUND, LP, NEW YO

Free format text: SECURITY INTEREST;ASSIGNOR:WSOU INVESTMENTS, LLC;REEL/FRAME:043966/0574

Effective date: 20170822

AS Assignment

Owner name: WSOU INVESTMENTS, LLC, CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:NOKIA TECHNOLOGIES OY;REEL/FRAME:043953/0822

Effective date: 20170722

AS Assignment

Owner name: WSOU INVESTMENTS, LLC, CALIFORNIA

Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:OCO OPPORTUNITIES MASTER FUND, L.P. (F/K/A OMEGA CREDIT OPPORTUNITIES MASTER FUND LP;REEL/FRAME:049246/0405

Effective date: 20190516