EP3884589A1

EP3884589A1 - Techniques for reducing quantization errors in electronically steerable antenna

Info

Publication number: EP3884589A1
Application number: EP19818283.4A
Authority: EP
Inventors: Dotan GOBERMAN; Doron Rainish
Original assignee: Satixfy UK Ltd
Current assignee: Satixfy UK Ltd
Priority date: 2018-11-19
Filing date: 2019-11-18
Publication date: 2021-09-29
Also published as: WO2020105040A1

Abstract

Beamforming techniques, and corresponding implementations, are disclosed, wherein digital data adapted by a beamforming process for transmission by a plurality of transmit channels of a ESA system is manipulated by a respective plurality of delta- sigma modulation processes for causing different modifications of at least a portion of the adapted digital data in each of the plurality of transmit channels. This way, variance can be introduced between signals transmitted by the transmit channels. The manipulated data of each transmit channel can be then converted into a corresponding analog domain signal for transmission thereof via the respective antenna element. The different modification obtained by manipulating the adapted digital data by the delta- sigma modulation processes can be configured to cause a reduction in the average correlation between the quantization noise signals transmitted from the respective antenna elements after the conversion into the analog domain, so that out of band quantization noise and errors can be substantially reduced at a receiver of the transmitted signals.

Description

TECHNIQUES FOR REDUCING QUANTIZATION ERRORS IN ELECTRONICALLY STEERABLE ANTENNA

TECHNOLOGICAL FIELD

The present invention is generally in the field of digital beamforming systems, and particularly of suppressing quantization errors in such systems.

BACKGROUND

Electronically Steerable Antenna (ESA) systems offer many advantages including electronic beam steering and scanning, optimized beam pattern with reduced sidelobes, and reduced power consumption and weight. True-time-delay (TTD) steering techniques are typically required for controlling operation of multiple antenna elements in the ESA system, while keeping the broad bandwidth of the antenna radiation and allowing large scan angle, so that efficient elemental vector summation (in the receive mode) or distribution (in the transmit mode) can be obtained, that is independent of frequency or angle of the transmitted or received signals.

Typical implementations of electronically steerable antennas are based on analog (RF) phase shifting. Hence the term Phased Array Antenna (PAA) is commonly used to describe ESA systems. In the following description the terms EISA and PAA are used interchangeably.

Analog ESA implementations suffer from several drawbacks, such as:

• due to implementation difficulties of the ESA systems TTD is almost never achieved in such analog implementations;

• analog phase shift units are typically non-accurate due to production variations;

• multi beam is very difficult to implement due to RF summation/splitting losses; and

• large antennas are difficult to implement due to the need of an accurate and low loss-routing.

Digital implementations of ESA systems don’t have the above drawbacks. However, digital implementations require a digital-to-analog converter (DAC) in the transmit path/channel of every antenna element of the array (and similarly an ADC in each receive path/channel). The DAC used in each transmit channel of the ESA system introduces quantization noise into the signal transmitted from the respective antenna element of the array. The analog signals generated by the DACs are transmitted simultaneously from the antenna elements of the array, and the quantization noise of all antenna elements coherently summated over the transmission medium (i.e., over the air), which produces out of band emission, deteriorates the signal-to-noise ratio (SNR) at the receiver, and can cause quantization errors at the receiving end. The effect of the quantization noise on SNR degradation becomes more problematic when the sampling resolution is low.

It should be noted that the antenna array gain does not substantially influence the out of band quantization noise level at the receiver antenna output since the quantization noises from the DACs of the transmit channels are highly correlated at that point.

Some solutions from the patent literature are briefly described below.

US 5,103,232 describes means of decorrelating phase quantization errors in a phased array radar antenna using digital randomization at each of the array elements to reduce peak steering errors and to reduce peak sidelobe levels of the antenna. A random phase adjust term is provided to each of the array's antenna elements which comprises a distributed controller (DC) co-located with a digital phase shifter. The distributed controllers are each programmed with a random phase adjust term which represents a phase shift adjustment statistically independent from element to element. The random phase adjust term is stored in a memory located in each distributed controller. The distributed controller drives each element's digitally controlled phase shifter in response to a beam steering command received over a serial line.

US 2015/365151 describes an antenna arrangement configured for digital beam forming of a transmit signal comprising; a number N>1 of DACs, each of the N DACs being arranged to receive one respective digital transmit signal component, and to convert and output an analog transmit signal component, each of the N DACs having a respective resolution below a resolution required to fulfill a regulatory radio requirement in an interchangeable antenna arrangement arranged for transmission by a single antenna element connected to a single DAC; and N antenna elements, each of the N antenna elements being configured to receive one respective analog transmit signal component and to transmit the analog transmit signal component as part of the digitally beam-formed transmit signal.

Reference to any prior art in the specification is not an acknowledgment or suggestion that this prior art forms part of the common general knowledge in any jurisdiction or that this prior art could reasonably be expected to be understood, regarded as relevant, and/or combined with other pieces of prior art by a skilled person in the art.

GENERAL DESCRIPTION

The present application provides techniques, and corresponding implementations, for substantially reducing/suppressing correlation of quantization noise signals in ESA systems utilizing digital beamforming. Digital beamforming provides various advantages in ESA systems, particularly due to the accuracy and flexibility obtained using TTD to implement the beam steering and scanning. However, digital beamforming is more susceptible to quantization errors, since each transmit channel of the ESA/phased array antenna system requires at least one DAC to convert the digital signals generated by the digital beamforming process into corresponding analog signals for RF transmission via the antenna elements (and similarly each receive channel requires at least one ADC).

In the transmit stage, for example, the quantization errors occur due to the high correlation between the quantization noise signals introduced by the DACs of the transmit channels of ESA systems, resulting in constructive interference of the quantization noise signals when the analog signals produced by the DACs are transmitted by the ESA, and therefore causing high out of band power emission. There are many advantages for operating ESA systems at low sampling resolutions (e.g., efficiency, simplified design, improved thermal properties, etc.), but on the other hand, as the bit length/depth of the DACs in the ESA system is reduced the maximal transmitted SNR becomes the quantization SNR (SNR_Q) of a single antenna element, which does not benefit from the array gain.

In the transmit path, quantization errors can be reduced/suppressed by introducing randomness to the signals to be transmitted by the ESA system, for example, by adding random low level noise signals to the signals supplied to each antenna element. Such solutions may require additional hardware means and generation of noise signals in each transmit channel of the ESA system and may decrease the SNR at the receiver.

The present application provides in some embodiments digital beamforming techniques configured to manipulate the signals digitally processed therein and introduce some level of variance between the digital signals prepared by the different transmit channels, and thereby cause for substantial decorrelation of the quantization noise signals introduced by the DACs of the ESA system i.e., without requiring generation of random noise signals. The de-correlation of the quantization noise signals induced by the transmit channels improves performance of the ESA system, since the beam-formed signal can be transmitted with uncorrelated distortion components, with zero mean value and without bias, to thereby substantially diminish error vector aggregation at the receiving end. One important effect of this approach is that out-of- band noise introduced by the ESA system is reduced, and thus the complexity and the cost of filtering this out-of-band noise is reduced.

Some of the digital beamforming techniques disclosed herein employ properties of elements utilized in digital beamforming processes to introduce the variance between the signals digitally processed by the different transmit channels of the ESA system, such that DACs of each transmit channel are fed by different variants of the signals, thereby reducing/suppressing correlation of the quantization noise from the DACs. This is achieved in some embodiments by utilizing delta-sigma (DS) modulation unit(s) in each transmit channel, and configuring each of the transmit channels to cause permutations in time to data generated by its respective DS modulation unit(s).

In some embodiments each transmit channel is configured to arrange the stream of data samples generated by its respective DS modulation process in a form of words, each word comprising one or more sequences of the data samples of predefined bit lengths i.e., each word of samples containing a predefined number of data samples bits outputted by the respective DS modulation process, and perform sample time permutations in at least one segment of each of the constructed words, to thereby affect permutation in time domain of the data outputted by the DS modulation process (also referred to herein as time permutation or sample time permutation). The digital beamforming process can be further configured to dynamically change for each transmit channel the permutation scheme(s) to be applied in one or more of the transmit channels. Optionally, the bit length of a segment in at least one transmit channel is a single bit, or greater than one bit.

For example, and without being limiting, at least one pseudo-random crossbar network can be generated for one or more of the transmit channels, and implemented in its beamforming process such that the digitally processed samples therein undergo the sample time permutations defined by the respective crossbar network before converted by the respective DAC into a corresponding analog signal.

In some embodiment the digital beamforming process can utilize at least one delta-sigma (DS) modulator in each transmit channel to oversample digital beamformed data, and/or reduce samples bit-depth, and/or for noise shaping. The samples generated by each DS modulator can be used to construct data words of a predefined length, and at least one segment of each word can be then permuted in time domain by the respective at least one pseudo-random crossbar network, before the data word is converted by the respective DAC into analog signals. In some embodiments the DS modulator of each transmit channel can be configured to implement loop filter having a slightly different transfer function of its filter unit.

Optionally, but in some embodiments preferably, the DS modulator of one or more of the transmit channels is operated with one or more initial conditions different from the initial conditions used in the DS modulator of the other transmit channels. Alternatively, or additionally, one or more DS modulators of certain transmit channels are initialized to operate with determined initial conditions at different point in time e.g., each DS modulator is initialized with same, or different, initial condition values, at a different time. Optionally, but in some embodiments preferably, at least one of the DS modulators is configured to exhibit chaotic behavior (e.g., complying with the Devaney definition of chaos). In this way, a desirable level of variance between the data supplied to the DACs of the transmit channels can be maintained throughout operation of the ESA system.

The term deterministic modification used herein to refer to manipulation of digital data that modify the data in a known, and optionally controlled, manner, without uncertainty about the modified data obtained i.e., the modified data obtained is generated by system components not involving randomness. Some of the deterministic modifications used in embodiments disclosed herein utilize samples time permutation schemes and/or delta-sigma processes. However, in some embodiments one or more of the parameters used for the deterministic modifications are parameters that can be generated utilizing random/pseudo-random number generation processes e.g., initial conditions, filter coefficients, permutation schemes, and suchlike. The use of such deterministic manipulations is utilized some embodiments to alter digital data processed in transmit channels of ESA systems in a controlled and adjustable manner for introducing a suitable/predefined level of variance between the transmit channels without corrupting the data signals to be thereby transmitted.

One inventive aspect of the subject matter disclosed herein relates to a beamforming method comprising causing modifications (e.g., deterministic modifications) to at least a portion of digital data to be transmitted by a plurality of transmit channels of a ESA system to thereby introduce variance between the signals transmitted by the transmit channels. The digital data may be prepared by performing digital beamforming in each of the transmit channels, to thereby adapt the digital data for transmission via a respective antenna element of the transmit channel.

The manipulating of the digital data to be transmitted is carried out in some embodiments using a delta-sigma modulation process in each transmit channel. The delta-sigma modulation process can be configured to add a constant value to an input of a quantization process thereof.

These modifications to the digital data to be transmitted by the transmit channels can be used for introducing variance between signals transmitted by the transmit channels. The method can comprise converting the manipulated data of each transmit channel to a corresponding analog domain signal for transmission thereof via the respective antenna element. The different modification obtained by manipulating the adapted digital data by the delta-sigma modulation can be configured to cause a reduction in the average correlation between the quantization noise signals transmitted from the respective antenna elements after the conversion into the analog domain. This way out of band quantization noise and errors can be substantially reduced at a receiver of the transmitted signals.

Optionally, but in some embodiments preferably, a transfer function of the delta- sigma modulation process is of a first order. The delta-sigma modulation process performed in each transmit channel can be configured this way for at least partially causing the different modification and the variance associated therewith. Optionally, the constant value added to quantization processes of at least some of the delta-sigma modulation processes is selected to be a different value. In some embodiments the constant value used in each transmit channel is selected from a set of numbers uniformly distributed within an output range of the respective delta-sigma modulation process. The method comprises in some embodiments defining a different transfer function for at least one, or for each, of the filter units of the delta-sigma modulation processes, to thereby at least partially cause the variance between the data supplied to the DACs of the transmit channels. Optionally, but in some embodiments preferably, the transfer function of the filter units is of a second order, or of a higher order. The method optionally comprises defining at least one different parameter of a transfer function of a filter unit for at least one of the delta-sigma modulation processes to thereby cause the variance between the digital signals of the different transmit channels.

Optionally, but in some embodiments preferably, a different noise transfer function is defined for at least one of the delta-sigma modulation processes of the transmit channels, to thereby cause the variance between the digital data of the transmit channels. For this purpose the method comprises in some embodiments determining at least one different parameter to at least one of the noise transfer functions.

The method comprises in some embodiments defining a same transfer function for the filter unit of each of the delta-sigma modulation processes, and defining at least one different parameter of the same transfer function of the filter unit in each of the transmission channels, to thereby at least partially cause the variance between the digital data of the transmit channels. The method can comprise defining for each of the delta- sigma modulation processes poles causing a band stop at a different frequency in each of the transmission channels.

Optionally, but in some embodiments preferably, a different phase Q value (e.g., as defined in the two equations below) is defined for a noise transfer function of at least one of the delta-sigma modulation processes, to at least partially cause the variance between the digital data of the transmit channels. The noise transfer function of the delta-sigma modulation processes is characterized in some embodiments by the expression -

(l-z^_1)(eⁱ⁹-z^_1)e^_i9.

Alternatively, the noise transfer function of the delta-sigma modulation processes is characterized by the expression -

1 -( 1 +2cos0)z^_1+( 1 +2cos0)z^-2- z^-3.

The method comprises in some embodiments defining at least one of the delta- sigma modulation processes to exhibit chaotic behavior. Initial conditions can be determined for at least one of the delta-sigma (DS) modulation processes. The term initial conditions used herein to refer to initial values of dynamic variables (also referred to as state variables) of a transfer function of the DS modulation process that affect their values and state at future times. In some embodiments each DS modulation process comprises at least one memory unit for storing state variables of a transfer function thereof, and the method comprises an initialization step, performed during system startup or after the system is reset, in which a specific sequence of input samples/values written into the memory cause the DS modulation to enter a certain modulator state. Optionally, the initial conditions are applied by resetting the memory of a filter of each DS modulation process, and then inputting thereinto specific input samples/values that will cause the required state of the DS modulation process, or alternatively by directly writing the input sample/values into the memories of the filters.

Alternatively, the method comprises defining the delta-sigma modulation processes to exhibit similar, or same, chaotic behavior/properties. In this alternative the method can comprise determining different initial conditions for each of the delta-sigma modulation processes. In another variant the method comprises defining the delta-sigma modulation processes to exhibit similar, or same, chaotic behavior, and initializing each the delta-sigma modulation processes with the same initial conditions at a different point in time. The initialization of at least one, or all, of the delta-sigma modulation processes with the determined initial conditions can be carried out periodically according to a defined initialization frequency of the system, or intermittently e.g., if instability of at least one delta-sigma modulation process is identified.

Optionally, the method comprises monitoring a state of at least one of the delta- sigma modulation processes, and adjusting the at least one delta-sigma modulation process whenever identifying that it is becoming unstable. For example, and without being limiting, if at least one chaotic DS modulator becomes unstable it is adjusted by the system to restore stability thereof. The adjusting can comprise at least one of the following: defining a different transfer function of a filter unit for the at least one delta- sigma modulation process, defining a different noise transfer function for the at least one delta-sigma modulation process, determining at least one different parameter for the noise transfer function of the at least one delta-sigma modulation process, defining a different phase Q value of the noise transfer function of the at least one delta-sigma modulation processes, determining different initial conditions for the at least one delta- sigma modulation process, and/or initializing the at least one delta-sigma modulation process with newly or previously determined initial conditions.

The method comprises in some embodiments determining for each of the transmit channels a size of the portion of the adapted digital data to be manipulated.

Optionally, but in some embodiments preferably, the different deterministic modification comprises applying in each transmit channel a different sample permutation in time domain to the at least a portion of the data to be transmitted. The method can thus comprise determining for each of the transmit channels a different time domain permutation scheme. Applying of the sample permutation may comprise constructing from the output of each delta-sigma modulation process a data word of a predefined size, partitioning each of the constructed words into a predetermined number of segments, and performing the sample order permutation in at least one of said segments of each word. Optionally, but in some embodiments preferably, a different sample order permutation scheme is applied to at least one of the following: the partitioned segments; and the constructed words.

Applying of the sample order permutation comprises in some embodiments constructing from a sample stream from each delta-sigma modulation process a predetermined number of sub-streams, partitioning each sub-steam into a predefined number of segments, applying a defined permutation scheme to each of the segments, and constructing from the permuted segments a permuted output sample stream.

The size of the portion of the adapted digital data to be manipulated can be determined based on at least one of: SNR conditions of the ESA system, an over sampling ratio of the system, and /or error correction capabilities of a receiver of the transmission.

Another inventive aspect of the subject matter disclosed herein relates to a beamforming system configured to process digital data for transmission by a ESA system and reduce quantization noise in signals thereby transmitted. The system comprises in some embodiments a plurality of transmit channels, each transmit channel associated with an antenna element of the ESA system and comprises a digital beamforming unit configured to process digital data to be transmitted by the ESA system e.g. , by affecting a phase shift (relative to the signals of the other transmit channels) and/or change/adjust signal amplitude and/or apply time delay thereto. A data manipulation unit is utilizing in some embodiments to apply a different modification (e.g., a deterministic modification) to at least a portion of the digital data processed by the digital beamforming unit, to thereby introduce variance between the digital data produced by the transmit channels.

Optionally, but in some embodiments preferably, the data manipulation unit comprises a delta-sigma modulator. The delta-sigma modulator used in each transmit channel can be configured for at least partially causing the different modification. Optionally, at least one of the delta-sigma modulators comprises a summation module configured to add a constant value to an input of a quantizer thereof. Thereby, variance between the data transmitted by the transmit channels can be introduced. A digital to analog converter can be used for converting the digital data modified by the data manipulation unit to a corresponding analog signal for transmission by the antenna element.

The delta-sigma modulator can define an oversampling ratio of the transmit channel. The delta-sigma modulator used in each transmit channel can be configured for at least partially causing the different modification.

The system comprises in some embodiments a control unit configured and operable to control operation of at least one of the data manipulation units. The control unit can be configured to determine an oversampling ratio for the plurality of transmit channels. The control unit can be also configured to determine the size of the portion of the adapted digital data to be manipulated in the at least one of the plurality of transmit channels based on the determined oversampling ratio.

The system comprises in some embodiment a delta-sigma setup module (e.g., provided in the control unit), which can be configured and operable to select a different constant value for the delta-sigma modulator. The delta-sigma setup module can be configured and operable to select the different constant values from a set of numbers uniformly distributed within an output range of the respective delta-sigma modulator. Optionally, the delta-sigma setup module is configured and operable to determine a different transfer function of a filter unit of each of the delta-sigma modulators and thereby at least partially cause the variance between the data of the transmit channels.

Optionally, the delta-sigma setup module is configured and operable to determine at least one different parameter of a transfer function of a filter unit of at least one of the delta-sigma modulators, and thereby at least partially cause the variance between the data of the transmit channels. In some possible embodiments the delta- sigma setup module is configured and operable to determine a different noise transfer function for at least one of the delta-sigma modulation processes, and thereby at least partially cause the variance between the data of the transmit channels. The delta-sigma setup module can be further configured to determine at least one different parameter of at least one of the noise transfer functions.

In some embodiments the delta-sigma setup module is configured and operable to determine a same transfer function for each of the delta-sigma modulators, and to determine at least one different parameter for at least one of the transfer functions, and thereby at least partially cause the variance between the data of the transmit channels. For this purpose the delta-sigma setup module can be configured to determine for each of the delta-sigma modulators poles causing a band stop at a different frequency i.e., to affect a different band-stop frequency in each delta-sigma setup module. In some embodiments the delta-sigma setup module is configured to determine a different phase Q value of a noise transfer function of at least one of the delta-sigma modulators.

Optionally, but in some embodiments preferably, at least one of the delta-sigma modulators is implemented as a chaotic delta-sigma modulator. The delta-sigma setup module can be configured and operable to determine initial conditions for at least one of the delta-sigma modulation processes. In some possible embodiments all of the delta- sigma modulators are implemented as chaotic delta-sigma modulators. The delta-sigma setup module can be thus configured and operable to determine different initial conditions for each of the delta-sigma modulators.

In another variant all of the delta-sigma modulators are implemented as chaotic delta-sigma modulators exhibiting similar, or same, chaotic behavior, and the delta- sigma setup module is configured and operable to initialize each of the delta-sigma modulators with the same initial conditions at a different point in time. Optionally, the control unit and/or the delta-sigma setup module, configured and operable to periodically or intermittently initialize at least one of the delta-sigma modulators with the determined initial conditions.

In some embodiments the control unit is configured and operable to monitor a state of at least one of the delta-sigma modulators and adjust the at least one delta-sigma modulator whenever identifying that it is becoming unstable. The control unit can be configured and operable to carry out at least one of the following when identifying that the at least one of the delta-sigma modulators is becoming unstable: define a different transfer function of a filter unit of the at least one delta-sigma modulator, define at least one different parameter of a transfer function of the filter unit of the at least one delta- sigma modulator, define a different noise transfer function for the at least one delta- sigma modulator, determine at least one different parameter for the noise transfer function of the at least one delta-sigma modulator, define a different phase Q value of the noise transfer function of the at least one delta-sigma modulator, determine different initial conditions for the at least one delta-sigma modulator, and/or initialize the at least one delta-sigma modulator with newly or previously determined initial conditions.

The data manipulation unit in at least one of the transmit channels can comprise a permutation unit configured and operable to apply a defined time domain samples permutation to the at least a portion of the digital data, and thereby at least partially cause the variance between the data of the transmit channels. Optionally, but in some embodiments preferably, the size of the data/samples permuted by the manipulation unit is determined based on the oversampling ratio defined by the delta-sigma modulator.

Alternatively, the data manipulation unit in each of the transmit channels comprises a permutation unit configured to apply a different time domain samples permutation scheme to the at least a portion of the digital data. The data manipulation unit in each transmit channel can be configured to construct from the output of its respective delta-sigma modulator a data word of a predefined size, partition the constructed word into a predetermined number of segments, and the manipulation unit can be configured to perform the sample time permutation to at least one of the segments. Optionally, but in some embodiments preferably, the manipulation unit is configured to perform a different sample time permutation in each of the segments.

The data manipulation unit is configured in some embodiments to construct from a sample stream from each delta-sigma modulator a predetermined number of sub streams, partition each sub-steam into a predefined number of segments, apply a different sample time permutation scheme to each of the segments, and construct from the permuted segments a permuted output sample stream.

The control unit comprises in some embodiments a permutation setup module configured and operable to determine for each of the transmit channels the portion size of the digital data to be modified by its respective data manipulation unit. The permutation setup module can be configured and operable to determine for each of the transmit channels the different permutation scheme used therein. Optionally, the permutation setup module is configured and operable to determine the portion size of the processed digital data to be manipulated based on at least one of: SNR conditions of the ESA system, an oversampling ratio of the system/transmit channel, and/or error correction capabilities of a receiver of the transmission.

Another inventive aspect disclosed herein relates to a data communication system utilizing the beamforming system/techniques disclosed hereinabove or hereinbelow.

Any of the preferable features described herein may be applied to any aspect of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to understand the invention and to see how it may be carried out in practice, embodiments will now be described, by way of non-limiting example only, with reference to the accompanying drawings. Features shown in the drawings are meant to be illustrative of only some embodiments of the invention, unless otherwise implicitly indicated. In the drawings like reference numerals are used to indicate corresponding parts, and in which:

Fig. 1 is a block diagram schematically illustrating an array antenna / electrical scanning antenna (ESA) implementation according to some possible embodiments;

Figs. 2A to 2D are block diagrams schematically illustrating use of delta-sigma (DS) modulation in transmit paths of a ESA system according to some possible embodiments, where Fig. 2A shows the ESA transmit paths, Fig. 2B and 2C show possible implementations of a DS modulator, and Fig. 2D illustrates possible elements of a delta-sigma setup module of the control unit;

Figs. 3A to 3D are block diagrams schematically illustrating use of delta-sigma (DS) modulation and samples permutations in transmit paths of a ESA system according to some possible embodiments, where Fig. 3A shows the ESA transmit paths, Fig. 3B and 3C show a possible sample-stream segmentation and permutation schemes, and Fig. 3D exemplifies carrying out permutation to sub-streams of samples from the delta- sigma (DS) modulator; and

Fig. 4 is a flowchart schematically illustrating a beamforming process according to some possible embodiments. DETAILED DESCRIPTION OF EMBODIMENTS

This application provides techniques and implementations of ESA systems configured to substantially reduce quantization errors typically introduced in the transmit paths of ESA systems, and to substantially reduce out of band emission of the transmitted signal. The techniques disclosed herein can be advantageously used in digital beamforming applications, and they are especially useful for multiple-beam forming as used in ESA systems.

The embodiments disclosed herein generally aim to introduce decorrelation between multiple transmit paths of a ESA system, and thereby substantially minimize, or prevent, the quantization noise from constructively interfering at a receiving end. This is achieved in some embodiments by manipulating the digital data processed in each transmit channel during the beamforming process to introduce some level of variance between the signals processed in the digital domain of each transmit channel for transmission by the ESA system. Accordingly, the disclosed embodiments provide architecture for the transmit paths/channels of digital beamformers, which can be utilized to reduce both the cost and the power consumption of conventional digital beamforming implementations.

One or more specific embodiments of the present disclosure will be described below with reference to the drawings, which are to be considered in all aspects as illustrative only and not restrictive in any manner. In an effort to provide a concise description of these embodiments, not all features of an actual implementation are described in the specification. Emphasis instead being placed upon clearly illustrating the principles of the invention such that persons skilled in the art will be able to make and use the digital beamforming techniques, once they understand their principles. This invention may be provided in other specific forms and embodiments without departing from the essential characteristics described herein.

For an overview of several example features, process stages, and principles of the invention, the digital beamforming examples illustrated schematically and diagrammatically in the figures are intended for transmit channels of ESA systems. These beamformers are shown as one example implementation that demonstrates a number of features, processes, and principles used to reduce/suppress correlation between quantization noise signals transmitted by the ESA system, but they are not limited to transmit channels, and also useful for other applications (e.g., phase shift beamforming) and can be made in different variations. Therefore, this description will proceed with reference to the shown examples, but with the understanding that the invention recited in the claims below can also be implemented in myriad other ways, once the principles are understood from the descriptions, explanations, and drawings herein. All such variations, as well as any other modifications apparent to one of ordinary skill in the art and useful in beamforming applications may be suitably employed, and are intended to fall within the scope of this disclosure.

The embodiments described hereinbelow refers to both transmit and receive ESA, however, for the sake of simplicity only a transmit direction will be described hereunder. Fig. 1 is a block diagram schematically illustrating a ESA system 10 comprising n transmit channels, TP1, TP2,..., TPn (where n> 1 is an integer), where each transmit channel TPi (where 1 <i<n is an integer) comprises a digital portion lOd and an analog portion 10a. The digital portion lOd of each transmit channel Tpi comprises a phase/amplitude (P/A) setting unit 11 configured to determine a phase shift for the signal lOg to be transmitted and its amplitude, a delay unit 12 configured to affect a time delay to the downstream propagation of the signal lOg along the transmit channel Tpi, an optional digital summation unit 13 can be used to add one or more additional signals Di e.g., received from other/external heamformers (not shown), and a DAC (D/A) 16 for converting the delayed digital signals into analog domain (10a) signals.

In this specific and non-limiting example the input signal lOg comprises an in- phase component I, and a quadrature component Q, but in other possible embodiments other, or no, angle modulation schemes can be utilized. The DAC 16 is thus shown in this example as comprising two DAC units, the DAC 16i for converting the delayed in- phase digital component of the input signal lOg, and the DAC 16q for converting the delayed quadrature digital component of the input signal lOg. Likewise, the other elements of the digital domain lOd can be implemented by two or more sub-elements, configured to operate on respective components of the input signal lOg.

The analog portion 10a of each transmit channel Tpi comprises a low pass filter (LPF) 35 configured to filter the analog signal outputted from the respective DAC 16, a mixer unit 17 for modulating the analog signal generated by the DACs 16 in a carrier signal lOr, in this specific and non-limiting example two mixers are used, mixer 17i for modulating the in-phase component (I), and mixer 17q for modulating the quadrature (Q) component, an analog sum ation unit 18 can be used to sum the modulated in- phase and quadrature components and feed the summation result to an amplifying unit (e.g., power amplifier - PA) 18 configured to amplify the modulated signals and feed them to the respective antenna element 19 of the ESA system 10.

A control unit 21 comprising one or more processors lOp and memories 10m can be used to control operation of each transmit channel Tpi by respective control signals Ci thereby generated.

Due to the non-continuous transformation of the DACs 16, quantization noise is inevitably added into the signal converted into the analog domain by the respective DACs 16 in each transmit channel Tpi. The system 10 can be adapted to decorrelate the quantization noise signals of the different transmit channels Tpi, and thereby substantially reduce/prevent constructive interference thereof, for example, by introducing a low level noise signal (e.g., white Gaussian noise) in the signals Di added to the delayed signals by the summation units 13.

The addition of the low level noise signals to the data processed in each transmit channel introduces variance in the data processed in the different transmit channels providing for the decorrelation. In some embodiments the control unit 21 comprises a noise generator module lOg configured and operable to generate the low level digital noise signals added to the delayed signals in each digital branch lOd by the respective summation unit 13 i.e., in the respective Di signal. Optionally, noise signal can be added to the data processed in each transmit channel by the respective summation unit 13 in other embodiments disclosed herein.

It is noted that though the invention is demonstrated in Fig. 1 utilizing quadrature modulation techniques, the invention is not limited to such modulation schemes, so different modulation schemes can be similarly used in the transmit channels

Tpi.

In some possible embodiments variance is introduced between the digital signals of each transmit channel TPi using delta-sigma (or sigma-delta) DACs, as demonstrated by the ESA system 30 shown in Fig 2A. The ESA system 30 shown in Fig. 2A is substantially similar to the ESA system 10 of Fig. 1, but further comprises in the digital branch lOd of each transmit channel TPi the up-sample filters (e.g., interpolators) 14 and the DS units 15, between the digital summation unit 13 and the DACs 16, and corresponding adjustments of the control unit 32. As in Fig. 1, the analog branch 10a of each transmit channel TPi comprises a respective low pass filter (LPF) 35. The low pass filter 35 can be configured to filter the analog signal outputted from the respective DAC 16 to remove the high frequency components introduced by the respective DS unit 15.

The control unit 32 can be accordingly adapted to include a DS setup module 31 configured and operable to set various parameters of the DS units 15, and/or monitor and/or adjust the operation of the DS units 15.

In this specific and non-limiting example the in-phase and quadrature signal components from the digital summation unit 13 of each transmit channel TPi undergo up-sampling in the up-sample filters to increases the sampling frequency Fs to N*Fs (where N>1 is a positive number), 14i and 14q respectively, and the up-sampled in- phase and quadrature components ( x _{( n )} in Fig. 2B) are then processed by the respective DS units, 15i and 15q. In a similar manner, the in-phase and quadrature analog signals generated by the respective DACs, 16i and 16q, are filtered by the respective LPF units, 35i and 35q.

Fig. 2B schematically illustrates a general form of a DS modulator 33 usable in the DS units 15 of the ESA system 30. The DS modulator 33 comprises a subtractor 33d configured to subtract an output sample y(n ) of the DS modulator from an input sample x(n ) of the DS modulator 33 (the subtraction result is referred to herein as comparison error), a loop filter unit 33s having a defined transfer function L (z) configured to accumulate comparison errors from the subtractor 33d, an adder 33a configured to add the input signal x(n) of the DS modulator 33 to the output of the loop filter unit 33s, and a quantizer (Qu) 33q configured to generate a sample-stream corresponding to the summation value produced by the adder 33a.

The DS modulator 33 is configured to produce a high rate sample-stream which average level represents the average level of the input signal x(n). In some possible embodiments the DS setup module 31 of the control unit 32 is configured to set various parameters of the DS modulators 15 to cause the DS modulators 15 of each transmit channel TPi to operate slightly different from the DS modulators 15 of the other transmit channels TP/ (where 1 <j¹i<n is an integer), and to thereby introduce pseudo randomness/variance in the values inputted to the DACs 16 of the different transmit channels TPi. For example, and without being limiting, the DS setup module 31 can be configured to use the control signals Ci" generated by the control unit 32 to provide the DS modulators 15 instructions to implement in each transmit channel TPi a loop filter unit 33s having a slightly different transfer function L(z) to thereby introduce some level of variability in the digital branch lOd of each transmit channel TP , and thereby de-correlate the quantization noise signals produced by the DACs 16 of the ESA system 30. The DS modulator 33 can be associated with a memory unit for storing various data associated with its operation, such as, but not limited to, transfer function and/or noise transfer function of the DS modulator, and/or phase values thereof, initial conditions of the DS modulator, and suchlike.

This way, the DS modulators 15/33, which can be respectively used for reducing the bit-depths of signals of the different respective transmit channels TPi, can be adjusted such that different (less-correlated) quantization noise signals will be introduced to the different signals outputted from the DACs 16.

The use of the DS modulator 15/33 can provide various advantages in the ESA system 30, such as, but not limited to, lower number of bits supplied to the DACs 16 (/.£., reduced bit-depth/resolution - compared to a non-oversampling DAC), and noise shaping (low frequency noise is pushed to high frequencies. This way noise signals can be further reduced by using higher order DS modulators 15/33 in the transmit channels TPi, while also reducing the bit-depth. In some possible embodiments the DS modulators 33 are at least 2^nd order modulators (i.e., having at least two integrators).

The serial connection of the up-sample filters 14, the DS modulators 33, and the DACs 16, forms a DS-DAC 34, which can be utilized to de-correlate quantization noise signals caused by the DACs 16, as will be explained hereinbelow in detail. The penalty of the usage of the DS-DACs 34 is that the out of band quantization noise is very high. As a result, the LPFs 35 should be designed for steeper and larger analog filtering of the out-of-band noise for each transmit channel TP , which may increase the cost and power consumption of the ESA system 30.

Fig. 2C schematically illustrates a possible implementation of a DS modulator 33' usable in the DS units 15 of the ESA system 30 shown in Fig. 2A. The DS modulator 33' is substantially similar to the DS modulator 33 shown in Fig. 2B, but further comprises the additional adder 33t between the adder 33a and the quantizer 33q. The additional adder 33t is configured to input to the quantizer 33q a summation of the output of the adder 33a with a constant number T;. Optionally, the DS modulator 33' can be implemented either as a real DS modulator, a complex DS modulator {i.e., wherein T; can be a complex number), a baseband DS modulator, or a bandpass DS modulator.

In some embodiments the DS modulator (15) of each transmit channel TPi is implemented by the DS modulator 33' shown in Fig. 2C. Optionally, but in some embodiments preferably, a different T; ( 1 <i<n) value is inputted to the DS modulator 33' of each transmit channel TPi i.e., Ti¹Tj for 1 <j<n and i¹j. The DS setup module 31 of the control unit 32 can be accordingly configured to generate a different T; value for the DS modulator 33' of each transmit channel TPi, and to communicate the generated T, value to the respective DS modulator 33' via the control signals Ci". Although the value T; used in the DS modulator 33' of each transmit channel TPi is a constant value, the DS setup module 31 can be configured to generate new T; values from time to time e.g., during system startup and/or due to changes in the communication channel.

One advantage of using in each transmit channel TPi a DS modulator 33' configured to add a constant value T; to the value supplied to the quantizer 33q of the DS modulator 33' is that it permits implementing the DS modulator 33' as a first order modulator, which considerably simplifies the implementation. Additionally, adding a constant value T; to the value supplied to the quantizer 33q of the DS modulator 33' of each transmit channel TPi can cause some delay/interference to the feedback loop of the DS modulator 33', which can contribute to the decorrelation of the analog signals generated by the respective DAC 16 of the transmit channel TPi. It was observed that adding the constant value T; to the value supplied to the quantizer 33q in each transmit channel TPi can cause a substantial reduction in the average correlation between the quantization noise signals transmitted from the antenna elements of the transmit channels TPi (after converted into analog signals by the DAC of the channel). It is noted that some correlation between the quantization noise signals associated with some of the antenna elements can be made negative, which generally can contribute to the reduction in the average correlation between the quantization noise signals transmitted from the antenna elements, but may as well introduce decorrelation.

The constant value T; is not limited to a specific range, and different values ranges can be selected per specific implementation requirements. In some embodiments the constant value T; is selected to fall within the output range of the DS modulator. For example, and without being limiting, if the output of the DS modulator is between -1 and 1, the value of T; can be accordingly selected within that same range i.e., -1<T«<1. Optionally, but in some embodiment preferably, the T; values are evenly distributed along the output range y_min to y_max of the DS modulator

By way of a non-limiting example, if the ESA system 30 have eight transmit channels TPi (i.e., n= 8 antenna elements) and the output of the DS modulator is between -1 and 1 (i.e., y_min=- l and y_max=l), then T;=( 1/8) -1 + (i-l)*(l-(-l))/8=- 0.875+0.25*(i-l) i.e., Ti=(l/8)-l=-0.875, T₂=(l/8)-l+(l/4)=-0.625, T₃=(l/8)-l+(l/2)=- 0.375, T₄=(l/8)-l+(3/4)=-0.125, TS=(1/8)-1+1=0.125, T₆=(1/8)-1+(5/4)=0.375,

T₇=(1/8)-1+(3/2)=0.625, T₈=(l/8)-l+(7/4)=0.875. It is however noted that the allocation of the T; values is not required to be descending/ascending with respect to the number of the transmit channel i i.e., the evenly distributed T; values can be randomly permuted before they are assigned to the transmit channels TPi.

The use of the DS modulators 33' in the ESA system 30 provided good results in computer simulation performed for transmit channels TPi utilizing non-tapered antenna elements (i.e., where equal power level is outputted by all antenna elements). It is however noted that though the use of tapered antenna elements in the transmit channels TPi performs well using T; values that are evenly distributed between the output range of the DS modulator, it was also observed that the allocation/order of the T; values to the transmit channels TPi is of substantial importance.

Optionally, at least one T; value can be selected to be outside the output range of the DS modulator. It is noted that for possible embodiments the T; values are not randomly selected, but rather set to be fixed deterministic values that can be changed from time to time according to transmit channel conditions and/or configuration and/or system requirements. Optionally the T; values can be linear or non-linear time dependent values generated using some predetermined time dependent function e.g., T«=/Kt).

Fig. 2D shows a delta-sigma setup module 31 of the control unit 32, according to some possible embodiments. In some embodiments the delta-sigma setup module 31 comprises a filter generator 31g configured to determine a transfer function (such as L (z) of the loop filter unit 33s shown in Fig. 2B) for each DS modulator 15 of the ESA system 30. Optionally, but in some embodiments preferably, the delta-sigma setup module 31 is configured to determine for each transmit channel TPi a DS modulator of the 2^nd, or higher order, comprising more feedback loops, and/or element organization, different from that shown in Fig. 2B. The loop filter generated for each transmit channel TPi can be transferred over the control signals Ci" for implementation in the respective transmit channels TPi. The configuration of the DS modulators 15 is typically carried out during off-line time periods (e.g., during system startup or after system reset), and in possible embodiments it is a onetime process after which parameters of the DS modulators 33 remain unchanged.

The DS modulators 15 can operate at either baseband or at IF (intermediate frequency) when IF section is implemented digitally, or in RF (radio frequency) when RF section is implemented digitally. The examples provided below (and above) are intended for baseband implementations. However, transferring these examples to the RF range is obvious to those skilled in the art.

The transfer function of a loop filter unit of a DS modulator can be generally expressed by the following equation -

Eq. 1

In some embodiments the delta-sigma setup module 31 comprises a random/pseudo-random parameters generator 31r configured to generate random/pseudo-random parameters (e.g., ao, ai,..., a_n, and bi,..., b_m) for each loop filter unit generated by the filter generator 31g for the DS modulators 15 of each transmit channel TPi. The randomly/pseudo-randomly generated parameters can be transferred to the transmit channel TPi by the control signals Ci", with the filter configurations generated by the filter generator 31g.

In some possible embodiments, each of the DS modulators 15/33 is designed with a different noise transfer function, to thereby de-correlate the quantization noise from the DACs 16 and randomize quantization errors. The de-correlation of the quantization noise signals induced by the transmit channels TPi improves performance of the ESA system 30, as its beam-formed transmit signal has uncorrelated distortion components, which substantially diminish quantization errors at a receiver end.

In some embodiments the delta-sigma setup module 31 comprises a noise transfer function generator 31f configured to generate a different noise transfer function (NTF) for the DS modulator 15/33 of each transmit channel TPi, and transfer the same over the control signals Ci" generated by the control unit 32 to each transmit channel TPi. At least 2^nd, or higher, order DS modulators 15/33 are used in some embodiments in the transmit channels TPi of the ESA system 30, respectively, whereby the 2^nd (or higher) order DS modulators 15/33 are configured with different parameters of their noise transfer functions such that different/non-correlated quantization noise signals are introduced thereby respectively.

There are several ways to design multiple DS modulators 15/33, each having a different NTF. For example, and without being limiting, choosing poles of the NTF that will lead to band stop at different frequency at each antenna element will result in less correlated noise between the different antenna elements 19. Additionally, or alternatively, a random/pseudo-random, phase Q can be added to the noise at each element as in Eq. 2 below. One possible example of such family of DS modulators is the second order sigma delta modulators having one zero at 1 and another zero at b^iq, on the 'Z'-plane. The noise transfer function obtained in this case can be expressed by the following equation -

and the randomization/variance is achieved in some embodiments by randomly/pseudo- randomly choosing the angle Q of the b^iq zero, for the DS modulators 15/33 of each transmit channel TPi, to reside within the following range -

Accordingly, in some possible embodiments, a different angle 6 _k (wherein 1 <k<n is an integer) is randomly/pseudo-randomly, chosen from the following uniform distribution:

However, any other suitable random/pseudo-random distribution, or selection scheme, of the angle 9_k can be similarly used.

The resulting loop filter unit obtained for the DS modulator 15/33 of transmit channel TP& can be expressed by the following equation:

Eq. 5

When using loop filter unit with order higher than one, it is possible to design multiple DS modulators 15/33, each having a different noise transfer function, by placing one or more poles of the loop filter unit to ensure in band noise reduction, while choosing the other poles such that stability and divergence between different elements is achieved. Optionally, but in some embodiments preferably, the same poles that ensure in band noise reduction are used for the DS modulators 15/33 of all transmit channels TP&. One possible non-limiting example of this technique, using third order DS modulator with one pole of the loop filter unit at one (1), and with two other complex conjugate pair of poles of the loop filter unit with negative real part, on the 'Z'-plane. The NTF in this case can be expressed by the following equation:

and the loop filter unit of the DS modulator 15/33 of the transmit channels TP& can be expressed by the following equation:

where the angles 9_k are chosen in the same way as in the previous example.

In some embodiments the delta-sigma setup module 31 comprises a computation module 31c configured and operable to receive the filter implementation generated by the filter generator 31g for each DS modulator 15/33, and/or its noise transfer function from the noise transfer function generator 31f, and/or the respective random/pseudo random parameters generated by the parameter generator 31r for the generated filter structure, and determine based thereon the poles and/or zeros of each DS modulator 15/33. A stability evaluation module 31s can be used to evaluate for each transmit channel TP&, based on the determined poles and/or zeros from the computation module 31c, if the generated filter unit and its randomly/pseudo-randomly generated parameters are acceptable for use therein and satisfy predetermined stability and/or noise reduction conditions of the ESA system.

The stability evaluation module 31s can be configured to monitor each DS modulator 15/33 to detect instability, and generate an alert signal 31x whenever instability is identified. For example, and without being limiting, instability can be identified by monitoring the average output value of the DS modulators 15/33 (y(n) in Fig, 2B), and verifying that it does not exceed some predefined instability threshold value. This monitoring can be separately performed in each DS modulator 15/33, or centrally by the evaluation module 31s.

In some embodiments the filter generator 31g, and/or the parameters generator 31r, is adapted to configure at least one of the transfer functions to cause the respective DS modulator 15/33 to exhibit chaotic behavior. This can be achieved in some possible embodiments by configuring the generated transfer functions with at least one integrator pole outside the 'Z'-plane unit circle. Alternatively, or additionally, chaotic behavior of at least one of the DS modulators 15/33 can be achieved by configuring the DS setup module 31 to provide that the defined transfer function L(z) is a non-minimum phase function (i.e., having all poles inside the 'Z'-plane unit circle) and having a number of zeros outside the 'Z'-plane unit circle complying with the order of the function.

Chaotic behavior of at least one of the DS modulators 15/33 can be achieved in some embodiments by configuring the delta-sigma setup module 31 to provide that the defined transfer function L(z) meets the requirements set in “The Sigma-Delta Modulator as a Chaotic Nonlinear Dynamical System” by Donald O. Campbell, Waterloo, Ontario, Canada, 2007 (section 5.4).

The use of one or more chaotic DS modulators in the transmit channels TPi of the ESA system decorrelates the quantization noise signals introduced by the DACs, and thereby prevents/reduce quantization errors. In some embodiments all of the DS modulators 15/33 are configured as chaotic DS modulators. For example, the DS modulator 15/33 of each transmit channel TPi can be configured to present a different chaotic behavior, to thereby introduce the variance at the input of DACs to decorrelate the quantization noise signals.

In some embodiments all of the DS modulators 15/33 are configured to implement a chaotic DS modulator having the same properties, but activated with different initial conditions (i.e., of state variables of the modulators). By activating each of the chaotic DS modulators with different initial conditions, the DS modulator of each transmit channel TPi introduces the variance at the input of DACs to decorrelate the quantization noise signals. This way the out of band noise of the DACs of the transmit channels TPi becomes highly non-correlated. Accordingly, the DS setup module 31 comprises in some embodiments an initial condition generator module 31i configured and operable to generate initial conditions for one or more of the DS modulators 15/33, and transfer the determined initial conditions to the one or more of the DS modulators 15/33 over the control signals Ci" generated by the control unit 32.

In some embodiments the DS modulator 15/33 of each transmit channel TPi is initialized at a different time for introducing the variance required to decorrelate the quantization noise signals. The DS setup module 31 thus comprises in some embodiments an initial condition scheduler module 31h configured and operable to determine for the DS modulator 15/33 of each transmit channel TPi a different initialization set-point time T/. The initial condition scheduler module 31h can be configured to schedule the initialization of the DS modulator 15/33 responsive to the state of the validation signal 31x from the evaluation module 31s.

In some embodiments the initial condition scheduler module 31h is configured to activate the initial condition generator module 31i to generate new initial conditions IC / for one or more of the DS modulators 15/33, and transfer the same over the control signals Ci", when the initialization is required. Accordingly, the initial condition generator module 31i can be configured to generate the same set of initial conditions IC/ for all of the DS modulators 15/33 (IC/=IC/, for 1 </,/<«), and transfer the set of initial conditions to each DS modulator 15/33 at a different time (T/¹Tj, for 1 £j¹i£n) according to signals/data 31d received from the initial condition scheduler module 31h. Alternatively, the initial condition generator module 31i can be configured to generate a different set of initial conditions ld¹lCj (1 £j¹i£n) for each of the DS modulators 15/33, and simultaneously transfer to each DS modulator 15/33 its set of initial conditions IC/, or at different times (T/¹Tj, for 1 £j¹i£ri), according to signals/data 31d received from the initial condition scheduler module 31h.

For example, and without being limiting, the delta-sigma setup module 31 can be configured and operable to set, or reset, the memory (not shown) of the digital filter unit of each DS modulator 15/33, at different points in time. The control unit 32 can be accordingly configured to generate for each transmit channel Tpi respective control signals Ci" to instruct setting the respective memory of the digital filters to store therein a defined set of initial conditions IC/ at a defined time T/ (where T i¹Tj for 1 £j¹i£ri), or to reset e.g., zero the memory of the sigma delta digital filter unit. In Some embodiments the control unit 32, and/or the delta-sigma setup module 31, configured and operable to periodically set or reset the memory of the digital filter unit of each DS modulator 15/33. In the embodiment shown in Fig. 3A, the ESA system 40 comprises permutation units 24, each configured to receive data samples accumulated from the stream of sample-stream outputted by a respective DS modulator 15, and apply a different permutation scheme thereto. In the following description each DS modulator 15 (15i and 15q), is referred to as outputting a stream of data samples, where each data sample outputted by the DS modulator 15 consists of a certain number of bits {i.e., each sample may comprise one or more bits). As shown in Fig. 3A, in this embodiment all of the data samples outputted by the DS modulators 15 are inputted to respective permutation units 24, as explained hereinbelow.

As shown in Fig. 3B, in some embodiments a L-sample word is constructed from the sample-stream output from each DS modulator 15, and each L-sample word is divided into K segments Si, S2,..., SK (where L, K> lare integers), each segment Si (where 1 <i<K is an integer) having a predefined number of bits i.e., having one or more bits. In some possible embodiments all of the segments S / are of the same size/length. Optionally, but in some embodiments preferably, the size/length of at least one of segments S / is different from the size/length of at least one other segment S _/ (where 1 £j£K is an integer and i¹j).

In some embodiments the number of samples L accumulated to construct the L- sample word is chosen to be around the oversampling ratio N (OSR - the ratio between the DS bitrate and the transmitted symbol rate), or smaller than the OSR. Accordingly, the sequence of L samples outputted from each DS modulator 15 is used to construct a L-sample word, and each of the constructed L-sample words is then divided into K segments, Si, S2,..., SK, which are not necessarily of the same length/size. Optionally, but in some embodiments preferably, the K segments, Si, S2,..., SK, are of the same length/size.

As illustrated in Fig. 3B, each of the constructed L-sample words is fed into a respective permutation unit 24 that may be comprised of a respective number of permutation modules Pi, P2,..., PK. In this embodiment each permutation module P; receives a respective segment S; of the L-sample word, applies to it a predefined permutation scheme, and outputs a respective segment S;' in which at least some portion thereof is a result of permutation applied to at least some portion of the input segment S/. The new word L' obtained from the permutation unit 24, comprised of the permuted segments S;', is then fed to the respective DACs (16 in Fig.3A) for conversion into respective analog signal. It is noted that K can be any divider of a rounded value of the OSR e.g., since the OSR (N) is not necessarily an integer, the number of segments K can be a divider of {_.N], where "|_ J" designates the floor rounding function i.e., [_N]=a*K, where a> 1 is an integer.

Fig. 3C schematically illustrates random/pseudo-random permutation scheme 48 performed by one of the permutation modules P; of the permutation units (24) to respective segments S; of the L-samples word constructed from the output of the respective DS modulator 15. As shown, each permutation module P; receives a respective segment S; and alters locations of two or more portions thereof, S and Sl^r) (where r¹t are positive integers, l<r< and 1 <t<M), within the received segment S;. Optionally, but in some embodiments preferably, the location of each portion S/^r) (1 <r<M) of the received segment S; is altered. This way, in the new segment S;' outputted from the permutation module P; locations of at least two portions, S/^ and S_/ ⁿ, of the received segment S; are altered/permuted.

Each transmit channel TPi in the ESA system 40 comprises two DS modulators 15i and 15q. Thus, in a ESA system comprising n transmit channels TPi there is a total of 2 n DS modulators (15i and 15q), the output of each DS modulator is fed to a respective permutation unit, 24i and 24q, each of which having K permutation modules P;, where each permutation module P; is assigned a random/pseudo-random permutation scheme 48, and each permutation scheme 48 is configured to randomly/pseudo- randomly permute a respective segment Si of a respective L-samples word constructed from the output of the DS modulator. The number of different possible assignments in this configuration is ( M\)^K (where '!' is factorial operation, and S designates the bit). For example, if L=OSR=64 bits word is divided at the output of each DS modulator 15 into K= 4 segments, each DS modulator 15 is assigned with 4 segments of order =16. The number of different possible assignments/permutation schemes is in this configuration is (16!)⁴=~1.9*10⁵³, so the number of antenna elements 19 in such embodiments practically is un-limited.

Fig. 3D shows a possible sample permutation technique wherein the stream of samples outputted by each DS modulator is divided into a predetermined number of Z

(where Z>2 is an integer) sample sub-streams, ss^{{ 1} ss^{2) . ss^(Z) by cyclically directing each sample produced by the DS modulator to a sub-stream is a subsequent manner. In this specific and non-limiting example an input sample switching device 37 is cyclically changed between states (1), (2),...,(Z), to direct each produced sample from the input sample stream 37i to a consecutive sub-stream, ss^{( l )}, ss^{2) . ss^(Z). For example, if Z=4, then the first sample stream ss^{( l )} will comprise sample numbers 1,5,9,13,17,21,25,29,33,37, the second sample stream ss^{2) will comprise sample numbers 2,6,10,14,18,22,26,30,34,38, and so on.

The L samples of each sub-stream ss ^/) (where \<q£L is an integer) are then partitioned into K segments S_9i, Sg2, . . . , S_9K, and each segment Sgk (where 1 <k<K is an integer) is then permuted by a respective permutation unit P _qk to yield a permuted segment S qk'. The output switching device 38 is configured to cyclically change between states (1), (2),...,(Z), in synchronization with the input sample switching device 37, to direct a permuted sample from the permutation units to the permuted output sample stream 38u. The permuted output sample stream 38u is supplied to a respective DAC 16 in the transmit channel.

In some embodiments, each segment Sgk (or S; of Fig.3C) comprises one or more samples from the respective DS modulator, and the permuted order of the samples from the respective permutation unit(s) are then fed to the respective DAC 16 in the transmit channel. The output switching device 38 can be thus configured to direct at least one permuted sample to the output sample stream 38u in each of the states (1), (2),..., (Z). Optionally, the input and output sample switching device, 37 and 38, are synchronized to simultaneously change into the same state (1), (2),..., (Z).

Fig. 4 is a flowchart schematically illustrating a beamforming process 50 carried out in each of the transmit channels (TPi) of the ESA system, according to some possible embodiments. The process 50 starts in receiving in step 51 the data to be transmitted by the ESA system. The received data is processed in step 52 to affect a respective phase and/or amplitude (PA) and/or delay (TTD), for transmission thereof by the respective antenna element (19) of the transmit channel, and in step 53 it is interpolated/up-sampled according to the over sample ratio (OSR). The over sample ratio value can be a predefined preset value of the system, or a value that is changed from time to time by the control unit of the system.

The beam-formed (TTD processed and/or phase-shifted and/or amplitude adjusted) and interpolated data then undergoes modification(s) in step 54 and 58, whereby at least some portion of the data undergoes a deterministic modification. As seen, the modification(s) stage can include only the DS modulation step 54, or both the DS modulation step 54 and the sample time permutation of step 58. The modification step(s) 54, or step 54 followed by step 58, are configured in some embodiments such that a different deterministic modification is applied to the processed data in each transmit channel (TPi), to thereby at least partially introduce the variance between the data of the different transmit channels.

In step 54, the DS modulation is applied to the processed data obtained in steps 52 and 53, where the DS modulation step performed in each transmit channel (TPi) is configured to use a different DS modulator, and/or a different loop filter unit L(z), and/or a different pseudo random phase Q value, and/or a different noise transfer function (NTF), and/or chaotic DS modulator simultaneously or periodically initialized by different initial conditions, or initialized by same initial conditions at different point in time, to at least partially cause the variance between the data supplied to the respective DAC in the transmit channel TPi. The DS modulated data can be then converted in step 55 into an analog signal, and thereafter filtered (by LPF) and transmitted, in steps 56 and 57, respectively.

Optionally, but in some embodiments preferably, sample time permutation is applied in step 58, in addition to the DS modulation of step 54 (shown by the double line arrows). The permutation step performed in each transmit channel (TPi) is configured to use a different crossbar scheme, and/or apply the permutation to a different portion of the data, and/or apply a sample time permutation to different length of the data, to at least partially cause the variance between the data supplied to the DACs of the different transmit channels TPi. Following the permutation of step 58 the modified data can be converted in step 55 into an analog signal, and thereafter filtered (by LPF) and transmitted, in steps 56 and 57, respectively.

It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable sub-combination. For example, the digital heamformer of a single transmit channel may include sample time permutation elements as discussed with reference to Figs. 3A to 3C, a modified DS modulator with up-sample filters as shown in Fig. 2A and/or 3A-3C. Alternatively, the individual features can be used separately. For example, the modified DS modulators can be used in a digital beamformer which does not incorporate sample time permutation elements.

Functions of the systems described hereinabove may be controlled through instructions executed by a computer-based control system. A control system suitable for use with embodiments described hereinabove may include, for example, one or more processors connected to a communication bus, one or more volatile memories (e.g., random access memory - RAM) or non-volatile memories (e.g., Flash memory). A secondary memory (e.g., a hard disk drive, a removable storage drive, and/or removable memory chip such as an EPROM, PROM or Flash memory) may be used for storing data, computer programs or other instructions, to be loaded into the computer system.

As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or computer program product. Accordingly, aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a "circuit," "module" or "system." Furthermore, aspects of the present invention may take the form of a computer program product embodied in one or more computer readable medium(s) having computer readable program code embodied thereon.

For example, computer programs (e.g., computer control logic) may be loaded from the secondary memory into a main memory for execution by one or more processors of the control system. Alternatively or additionally, Computer programs may be received via a communication interface. Such computer programs, when executed, enable the computer system to perform certain features of the present invention as discussed herein. In particular, the computer programs, when executed, enable a control processor to perform and/or cause the performance of features of the present invention. Accordingly, such computer programs may implement controllers of the computer system. In an embodiment where the invention is implemented using software, the software can be stored in a computer program product and loaded into the computer system using the removable storage drive, the memory chips or the communications interface. The control logic (software), when executed by a control processor, causes the control processor to perform certain functions of the invention as described herein. In another embodiment, features of the invention are implemented primarily in hardware using, for example, hardware components such as application specific integrated circuits (ASICs) or field-programmable gated arrays (FPGAs). Implementation of the hardware state machine so as to perform the functions described herein will be apparent to persons skilled in the relevant art(s). In yet another embodiment, features of the invention can be implemented using a combination of both hardware and software.

As described hereinabove and shown in the associated figures, the present invention provides digital heamformers configured to reduce/suppress correlation of quantization noise signals, and related methods. While particular embodiments of the invention have been described, it will be understood, however, that the invention is not limited thereto, since modifications may be made by those skilled in the art, particularly in light of the foregoing teachings. As will be appreciated by the skilled person, the invention can be carried out in a great variety of ways, employing more than one technique from those described above, all without exceeding the scope of the claims.

Claims

CLAIMS:

1. A beamforming method comprising:

applying in each one of a plurality of transmit channels of a ESA system a respective delta-sigma modulation process for manipulating digital data adapted by a beamforming process for transmission via said plurality of transmit channels, to thereby cause a different modification of at least a portion of the adapted digital data in each of said plurality of transmit channels and introduce variance between the digital data produced by the plurality of transmit channels; and

converting the manipulated data of each transmit channel to a corresponding analog signal for transmission thereof via said respective antenna element,

whereby said different modification caused by said respective delta-sigma modulation process is causing de-correlation of quantization noise signals introduced in said transmission by said conversion.

2. The method of claim 1 comprising adding a constant value to inputs of quantization processes of the delta-sigma modulation processes.

3. The method of claim 2 wherein transfer functions of the delta-sigma modulation processes is of a first order.

4. The method of any one of claims 2 and 3 comprising selecting a different constant value for adding to each input of the quantization processes of the delta-sigma modulation processes, said different number selected from a set of numbers uniformly distributed within an output range of the respective delta-sigma modulation process.

5. The method of any one of the preceding claims comprising determining size of the portion of the adapted digital data to be manipulated in the at least one of the plurality of transmit channels based on an oversampling ratio.

6. The method of any one of the preceding claims comprising defining a different transfer function for at least one filter, or for each filter, of the respective plurality delta- sigma modulation processes.

7. The method of claim 6 wherein the transfer function is of a second order, or of a higher order.

8. The method of any one of the preceding claims comprising defining at least one different parameter of a transfer function for at least one of the respective plurality of delta-sigma modulation processes.

9. The method of any one of the preceding claims comprising defining a different noise transfer function for at least one of the respective plurality of delta-sigma modulation processes.

10. The method of claim 9 comprising determining at least one different parameter for at least one of the noise transfer functions.

11. The method of any one of the preceding claims comprising defining for each of the respective plurality of delta-sigma modulation processes poles causing a band stop at a different frequency in each of the plurality of transmission channels.

12. The method of any one of the preceding claims comprising defining a different phase value of a noise transfer function of at least one of the respective plurality of delta-sigma modulation processes.

13. The method of any one of the preceding claims comprising defining a second order delta-sigma modulation processes for each of the plurality of sigma-delta modulation processes having one zero at 1 and another zero at b^iq, on the 'Z'-plane.

14. The method of any one of claims 12 or 13 wherein the noise transfer function of the respective plurality of delta-sigma modulation processes is characterized by the expression -

(l-z-^!Xe^-z-¹)^.

15. The method of claim 14 wherein a noise transfer function of the respective plurality of delta-sigma modulation processes is characterized by the expression

1 -( 1 +2cos0)z^_1+( 1 +2cos0)z ²-z^-3 .

16. The method of any one of the preceding claims comprising defining at least one of the respective plurality of delta-sigma modulation processes to exhibit a chaotic behavior.

17. The method of claim 16 comprising determining initial conditions for at least one of the respective plurality of delta-sigma modulation processes.

18. The method of claim 16 comprising defining all of the respective plurality of delta-sigma modulation processes to exhibit a similar chaotic behavior, and determining different initial conditions for each of the respective plurality of delta-sigma modulation processes.

19. The method of claim 16 comprising defining all of the respective plurality of delta-sigma modulation processes to exhibit a similar chaotic behavior, and initializing each of the respective plurality of delta-sigma modulation processes with the same initial conditions at a different time.

20. The method of any one of the preceding claims comprising monitoring at least one of the respective plurality of delta-sigma modulation processes, and adjusting said at least one delta-sigma modulation process whenever identifying that it is becoming unstable.

21. The method of claim 20 wherein the adjusting comprises at least one of the following: defining a different transfer function for the filter of at least one of the respective plurality of delta-sigma modulation process, defining at least one different parameter of a transfer function of the filter of at least one of the respective plurality of delta-sigma modulation process, defining a different noise transfer function for the at least one of the respective plurality of delta-sigma modulation process, determining at least one different parameter for the noise transfer functions of the at least one of the respective plurality of delta-sigma modulation process, defining a different phase Q value of the noise transfer function of the at least one of the respective plurality of delta- sigma modulation processes, determining different initial conditions for the at least one of the respective plurality of delta-sigma modulation process, and/or initializing the at least one of the respective plurality of delta-sigma modulation processes with newly or previously determined initial conditions.

22. The method of claim of any one of the preceding claims comprising determining for each of the plurality of transmit channels a size of the portion of the adapted digital data to be manipulated.

23. The method of any one of the preceding claims wherein the different modification comprises applying in each of the plurality of the transmit channels a different sample order permutation to the at least a portion of the data to be transmitted.

24. The method of claim 23 wherein the applying of the sample order permutation comprises constructing from the output of each of the respective plurality of delta-sigma modulation processes a data word of a predefined size, partitioning each of the constructed words into a predetermined number of segments, and performing the sample order permutation in at least one of said segments of each data word.

25. The method of claim 24 comprising applying a different sample order permutation scheme in at least one of the partitioned segments of the constructed data words.

26. The method of claim 23 wherein the applying of the sample order permutation comprises constructing from a sample stream from each one of the respective plurality of delta-sigma modulation processes a predetermined number of sub-streams, partitioning each sub-steam into a predefined number of segments, applying a defined permutation scheme to each of said predefined number of segments, and constructing from the permuted segments a permuted output sample stream.

27. The method of any one of claims 22 to 26 comprising determining the size of the portion of the adapted digital data to be manipulated based on at least one of the following: SNR conditions of the ESA system, an oversampling ratio of the transmit channel, and error correction capabilities of a receiver that receives the transmission.

28. A beamforming system configured to process digital data for transmission by a ESA system and reduce quantization noise in signals thereby transmitted, the system comprising a plurality of transmit channels, each of said plurality of transmit channels associated with a respective antenna element of said ESA and comprises: a respective beamforming unit configured to process digital data to be transmitted by the ESA system for affecting a phase difference and/or amplitude adjustment and/or time delay thereto;

a respective data manipulation unit comprising a delta-sigma modulator adapted to apply a different modification to at least a portion of the processed digital data from said respective digital beamforming unit and thereby introduce variance between the digital data produced by the plurality of transmit channels; and

a respective digital to analog converter for converting the modified digital data from said respective data manipulation unit into a corresponding analog signal for transmission by said respective antenna element.

29. The system of claim 28 wherein at least one of the delta-sigma modulators comprises a summation module configured to add a constant value to an input of a quantizer thereof.

30. The system of claim 29 comprising a control unit configured and operable to determine a size of the portion of the adapted digital data to be manipulated in the at least one of the plurality of transmit channels based on an oversampling ratio.

31. The system of any one of claims 28 to 30 comprising a delta-sigma setup module configured and operable to determine at least one of the following: a different constant value for each of the respective delta-sigma modulators; size of the portion of the adapted digital data to be manipulated in the at least one of the plurality of transmit channels based on an oversampling ratio; a different transfer function for a filter unit of each one of the respective plurality of delta-sigma modulators and thereby at least partially cause the variance between the data of the plurality of transmit channels; at least one different parameter of a transfer function of the filter unit for at least one of the respective plurality of delta-sigma modulators; a different noise transfer function for at least one of the respective plurality of delta-sigma modulators; at least one different parameter to at least one of the noise transfer functions; poles for the noise transfer function of each one of the respective plurality of the delta-sigma modulators for causing a band stop at a different frequency; a different phase Q value of the noise transfer function of at least one of the respective plurality of delta-sigma modulators.

32. The system of any one of claims 28 to 31 wherein at least one of the respective plurality of delta-sigma modulators is implemented as a chaotic delta-sigma modulator.

33. The system of any one of claims 31 and 32 wherein the delta-sigma setup module is configured and operable to determine initial conditions for at least one of the respective plurality of delta-sigma modulators.

34. The system of claim 33 wherein all of the respective plurality of delta-sigma modulators are implemented as chaotic delta-sigma modulators exhibiting similar chaotic behavior, and wherein the delta-sigma setup module is configured and operable to determine different initial conditions for each one of the respective plurality delta- sigma modulators.

35. The system of claim 34 wherein all of the delta-sigma modulators are implemented as chaotic delta-sigma modulators exhibiting similar chaotic behavior, and wherein the delta-sigma setup module is configured and operable to initialize each of the respective plurality of delta-sigma modulators with the same initial conditions at a different point in time.

36. The system of any one of claims 34 and 35 wherein the control unit is configured and operable to periodically or intermittently initialize at least one of the respective plurality of delta-sigma modulators with the determined initial conditions.

37. The system of any one of claims 28 to 36 wherein the data manipulation unit in at least one of the plurality of transmit channels comprises a permutation unit configured to apply a defined sample time permutation to the at least a portion of the digital data.

38. The system of any one of claims 28 and 37 wherein the respective data manipulation unit in each one of the plurality of transmit channels comprises a permutation unit configured to apply a different permutation scheme to the at least a portion of the digital data.

39. The system of claim 38 wherein the respective data manipulation unit is configured to construct from the output of its respective delta-sigma modulator one of the following: a data word of a predefined size, partition the constructed word into a predetermined number of segments, and wherein the manipulation unit is configured to perform a different sample time permutation to each of said segments; or a predetermined number of sub-streams, partition each sub-steam into a predefined number of segments, apply a different sample time permutation scheme to of the segments, and construct from the permuted segments a permuted output sample stream.

40. A data communication system comprising the beamforming system of any one of claims 28 to 39.