Accelerated keying for optical wireless communication systems

Abstract

The invention relates to a binary signaling scheme for use in a transmitter comprising a semiconductor radiation source, in which the drive current is modulated by switching it on and off, and thereby also the radiation output level. However, the modulation of the drive current is adapted not to wait for the radiation response to follow a complete rise or fall curve, but to switch between two (or a larger number of) selected radiation output levels. During a prolonged sequence of zeros (or ones), the drive current is switched to keep the radiation output level between a certain minimum and maximum level. This range of radiation output levels is chosen adaptively in dependence on the characteristics of the drive current, the optical transmission channel, the radiation source and / or the radiation detector.

Description

Technical Field

[0001] This invention relates to the field of signal transmission in optical communication networks, and more particularly to a method for driving a semiconductor radiation source at a transmitter, a transmitter for use in an optical communication system, and optical communication systems, such as, but not limited to, LiFi networks, for use in a variety of applications in homes, offices, retail, hotels, and industries. Background Technology

[0002] Optical wireless communication (OWC) systems such as LiFi networks (compared to WiFi networks) enable mobile user devices (hereinafter referred to as endpoints (EPs))—such as laptops, tablets, and smartphones—to wirelessly connect to the internet. WiFi uses radio frequencies to achieve this, but LiFi uses the spectrum, enabling unprecedented data transmission speeds and bandwidth. Furthermore, it can be used in areas susceptible to electromagnetic interference. An important point to consider is that wireless data is not just what our traditional connectivity devices require. Today, televisions, speakers, headphones, printers, virtual reality (VR) goggles, and even refrigerators use wireless data to connect and perform vital communications. Radio frequency (RF) technologies like WiFi are exhausting the spectrum supporting this digital revolution, and LiFi can help drive the next generation of immersive connectivity.

[0003] Based on modulation, any suitable light sensor can be used to detect information in the coded light. This could be a dedicated photocell (point detector), a photocell array possibly with lenses, a reflector, a diffuser with a phosphor converter, or a camera comprising an array of photocells (pixels) and lenses for forming an image on the array. For example, the light sensor could be a dedicated photocell included in a dongle inserted into the endpoint, or the sensor could be a general-purpose (visible or infrared) camera at the endpoint or an infrared detector originally designed for, for example, 3D face recognition. Either way, this allows applications running on the endpoint to receive data via light.

[0004] As an example, on-off keying (OOK) is an attractive modulation method widely used in fiber optic and OWC systems based on semiconductor transmitters or emitters such as laser diodes or light-emitting diodes (LEDs). In fact, based on a general understanding of communication principles, for a given power budget, the highest throughput can be achieved through two-level modulation. This understanding is based on interpreting communication signals as points in a multi-dimensional space. The use of multi-level signals requires M = 2. mM bits are transmitted using a signal level, and all these levels must be separated by a minimum distance to make the signal robust to noise. Therefore, avoiding the use of multiple bits per symbol is the most energy-efficient approach. However, using only two bits per symbol requires higher signal bandwidth.

[0005] International patent application WO 2017 / 190998 A1 discloses a light source for use in a visible optical communication system, which applies on / off keying modulation to a data sequence to embed the data signal as an on / off keying amplitude modulation of the data signal. To accommodate PWM dimming, D1 proposes synchronizing the data and dimming signals, and then adding the PWM dimming signal and the data signal to generate a drive signal for the LED light source. The resulting drive signal is provided to the LED light source, thereby embedding the data sequence into the light output of the light source.

[0006] Light sources (especially LEDs) have limited bandwidth. Junction capacitance acts as a low-pass filter. Secondly, the phosphors used in blue photon conversion white LEDs contribute to this low-pass behavior. The same applies to detectors, particularly those with large apertures and wide opening angles required in wireless optical communications. Bandwidth limitations can also occur in optical fibers, especially in multimode plastic fiber (POF). These POFs can be attractive for short-range links, such as connecting wireless transmitters in homes or apartments, or in office spaces in factory lobbies. In some applications, such POFs even combine signals transmitted wirelessly with wireless optical links, potentially involving electrical or optical amplifiers. All these effects contribute to the phenomenon of optical channels limiting the bandwidth of modulated signals.

[0007] Due to this limited bandwidth, today's LED-based LiFi systems typically use Orthogonal Frequency Division Multiplexing (OFDM) with multiple bits per symbol, which is more spectrally efficient. However, OOK, with only two signal levels, is more energy-efficient than OFDM. Furthermore, OFDM requires highly linear systems with power-hungry amplifiers and complex signal processing.

[0008] OOK modulation has another advantage: the driver can be implemented using fast switching, making it very power-efficient and compact. However, semiconductor transmitters / emitters (such as LEDs, lasers) or receivers / detectors (such as photodiodes) may limit the available bit rate due to the small opening in the receiver's eye diagram, making detection more difficult.

[0009] Due to the physical properties of semiconductor light sources (such as LEDs or laser diodes), it takes a certain amount of time to reach the steady-state low or high value of the binary drive signal. Laser diodes are faster, but their rise and fall times can still be limiting factors. If the bit rate is greater than, for example, twice the bandwidth of the semiconductor light source, inter-symbol interference (ISI) occurs. Summary of the Invention

[0010] One object of this invention is to provide a driving scheme for modulating a semiconductor light source at an increased bit rate exceeding 3 dB bandwidth of the light source, thereby penetrating into a region where severe crosstalk and / or eye closure of the eye diagram would otherwise occur (without using the measures disclosed herein). However, the invention aims to ensure that the bit error rate remains sufficiently low. Furthermore, the invention aims to address the properties of optical channels, which differ from the common situation in radio frequency (RF) systems where the modulation bandwidth is typically strictly limited to avoid overflow into adjacent channels, thus requiring steep filters. In intensity-modulated optical communications, the problem of overflow into adjacent wavelengths is generally not a primary design criterion, and in this respect, it differs from RF. However, for optical communications, particularly if LEDs are used, the aforementioned low-pass filtering mechanism results in a relatively gentle roll-off above a certain frequency, and therefore bandwidth above the 3 dB attenuation frequency can still be used for communication. Therefore, it is proposed to use modulation frequencies significantly higher than 3 dB bandwidth to achieve better throughput in OWC.

[0011] This objective is achieved through transmitters, optical communication systems, methods, and computer program products.

[0012] According to a first aspect, a transmitter is provided for generating a radiated signal in an optical communication system, the transmitter including means for driving a semiconductor radiation source, the means comprising: A modulator configured to modulate a drive signal provided to a radiation source with a data sequence to generate at least two levels of drive current through the radiation source, the at least two levels of drive current defining a radiation output range between an upper radiation output level and a lower radiation output level of the radiation source during the transmission of the data sequence; and The controller is configured to: determine the transmission quality of the radiation output at the transmitter using at least one of a feedback signal indicating the radiation output level of the radiation source and feedback information received from the receiver of the radiation output; determine an upper target level and a lower target level that define a reduced radiation output range based on the transmission quality of the radiation output; and adaptively control a modulator to control a drive current, and switch the drive current at a timing determined by the reduced radiation output range between the predetermined upper target level and the predetermined lower target level.

[0013] The radiation output range can be reduced by intentionally switching the current to avoid reaching a steady state instead of allowing the radiation source to reach a steady state.

[0014] According to a second aspect, a method is provided for driving a semiconductor radiation source at a transmitter used to generate a radiation signal in an optical communication system, the method comprising: The driving signal provided to the radiation source is modulated by a data sequence to generate a driving current of at least two levels through the radiation source, the driving current of at least two levels defining the radiation output range between the upper radiation output level and the lower radiation output level of the radiation source during the transmission of the data sequence. The transmission quality of the radiation output at the transmitter is determined using at least one of feedback information received from the receiver of the radiation output and a feedback signal indicating the radiation output level of the radiation source. The upper and lower target levels, which define the reduced radiation output range, are determined based on the transmission quality of the radiated output; and The drive current is adaptively controlled, and the timing of the drive current is switched as determined by the reduced radiation output range between the upper target level and the lower target level.

[0015] Therefore, due to the low-pass filtering effect of the semiconductor radiation source, the proposed driving scheme functions as a pre-compensation for the limited bandwidth on the transmitter side. This is achieved by driving the semiconductor radiation source with a higher drive current than would be required to reach the desired radiation level, but limiting the oscillation or range of the radiation output by lowering the upper limit of the effectively usable radiation output value and / or raising the lower limit of the effectively usable radiation output value. Thus, the semiconductor radiation source can be driven within maximum and / or minimum (even potentially negative) drive currents while modulating its radiation output with a limited current range that is a subrange of the total operating range of the semiconductor radiation source.

[0016] Typically, the data sequence used to generate drive currents of at least two levels will be a binary data sequence, although these aspects are not limited to this.

[0017] The drive current can then be controlled using either voltage or current control mechanisms. Preferably, the drive current used is the minimum / maximum drive current rating of the radiation source, but it may be exceeded when possible. In the latter case, precautions may be necessary to prevent damage to the radiation source. Due to modulation speed, the duration of the pulse applied to the radiation source will be short (typically faster than 1 MHz), and given the reduced radiation output range, exceeding the minimum / maximum drive current rating may be feasible without damaging the radiation source.

[0018] Therefore, this method allows for rapid modulation by driving a semiconductor source between selected discrete radiated output levels within a subrange. This subrange can be selected within a steeper slope range of the semiconductor source's response to the drive pulse, thereby reducing the asymmetry of the rise and fall slopes. Furthermore, due to the reduced radiated output range, a higher current can be used to drive the semiconductor source more quickly, while the bit duration can be set equal to the time taken to reach an output level sufficient to distinguish the logic states in the output signal. When that point is reached, the next bit time can begin. Thus, the actual alternating sequence of binary values ​​(e.g., zero and one) determines the achievable bit rate. For other sequences with more than one consecutive binary value, the drive current can be controlled to maintain a predefined range-limited radiated output level. Holding (on or off) the drive current can push the radiated output level out of its range, thus hindering the system's ability to reach a target radiated output level corresponding to a changed symbol (bit) value in a timely manner. The latter corresponds to inter-symbol interference. For example, during a sustained logic one, the driver can switch the current in an on / off mode to ensure the radiated output level remains near its range limit.

[0019] The transmission quality of the radiated output can be determined based on feedback information received from the receiver at the radiated output. Therefore, the upper and lower target levels of the radiated output can be controlled via drive signals based on the actual reception conditions at the receiver.

[0020] The rate of the drive signal and at least one of the maximum and / or minimum permissible output radiation level or output light range of the radiation source can be determined based on transmission quality. Therefore, the transmission speed of the radiation source can be adapted to the transmission quality of the radiation channel. For example, if the transmission quality increases (noise level decreases), it indicates that the channel transfer function is advantageous, the distance (range) between the maximum and minimum permissible output radiation levels (upper target level and lower target level) can be reduced (smaller eye opening), allowing the target level to be reached more quickly, and the switching rate of the increased drive current (i.e., bit rate) will increase, and vice versa. Conversely, when the transmission quality degrades below a threshold, the reduced range can be extended, resulting in a lower speed but with improved transmission quality.

[0021] Alternatively, a feedback signal indicating the radiation output level or other transmission quality indication of the radiation source can be received at the modulator, and the switching of the drive signal can be controlled based on the feedback signal. This allows the switching time of the drive signal to be directly adjusted in response to the radiation output level or other transmission quality indication of the radiation source.

[0022] As a trade-off between achieving higher transmission quality and reaching lower output levels in a shorter time period, the capacitors of semiconductor radiation sources do not need to be fully discharged. Thus, semiconductor light sources can be modulated at increased bit rates beyond regions where severe crosstalk and / or eye diagram closure occur, while maintaining a sufficiently low bit error rate.

[0023] Various methods can be used to establish the transmission quality of the radiated output. For example, the measurement / metric of transmission quality can be determined locally at the transmitter side, thus explicitly considering the output characteristics of (one or more) transmitters, but not the channel transmission characteristics. This may be sufficient, as the quality assessment and determination of the reduced range can take into account known constraints of the optical channel, such as typical communication distances, and / or known characteristics of the optical front-ends used (output and / or input). Alternatively, the measurement / metric of transmission quality can be determined at the receiver side, thus explicitly incorporating the channel transmission function. Monitoring transmission quality at the transmitter side may involve analyzing the radiated output level of the radiation source. A basic approach would be to monitor the voltage above the radiation source, particularly at the expected sampling time. Given a known reduced output range, this voltage can provide feedback on good quality. The voltage above a semiconductor radiation source is an indicator of the junction charge of the radiation source and therefore an indicator of the radiated output level. More advanced methods could attempt to further characterize the deviation of the symbol's voltage curve from the symbol's expected output curve and use this as a quality measurement / metric to also consider the voltage level outside the sampling time. Furthermore, various options for establishing transmission quality measurements / metrics on the receiver side are described in the options below.

[0024] Note that the proposed driving scheme also works well for transmission over optical fibers (e.g., polymer optical fibers (POF)), which allows for small and therefore fast detectors, where transmitter limitations are relevant. Therefore, higher bit rates can also be achieved for semiconductor light sources (e.g., LEDs) via optical fibers.

[0025] According to a third aspect, an optical communication system is provided, comprising a transmitter according to the third aspect and a receiver for receiving radiated signals generated by the transmitter.

[0026] According to a fourth aspect, a computer program product is provided, which includes code means for generating the steps of the method described in the second aspect when run on a controller device.

[0027] According to the first option of any of the first to fourth aspects, the upper target level and the lower target level can be maintained by switching the drive signal at a rate higher than the rate of the data sequence, which will be converted into a bit rate if the data sequence is a binary data sequence.

[0028] According to the second option of any one of the first to fourth aspects, which can be combined with the first option, the data sequence is a binary data sequence, and the binary switching mode can be used as a drive signal during the bit duration of the binary data sequence, wherein the binary switching mode is determined by a combination of the current bit value and at least one previous bit value of the binary data sequence. Therefore, the drive signal can be switched in sub-bit timings to ensure that the high and low levels of the radiated output can be maintained, and thus the range of the radiated output can be maintained.

[0029] According to the third option of any of the first to fourth aspects, which can be combined with any of the first to second options, when the transmission quality of the radiated output is based on the feedback signal, the feedback signal can be compared with the inversion lower limit and the inversion upper limit, and the drive signal can be turned off when the feedback signal has reached the inversion upper limit and / or the drive signal can be turned on when the feedback signal has reached the inversion lower limit.

[0030] According to the fourth option of any of the first to fourth aspects, which can be combined with any of the first to third options, a short-circuit or negative drive current can be applied to the radiation source during the off state of the drive signal. This increases the fall time of the radiation output from high to low level.

[0031] According to the fifth option of any of the first to fourth aspects, it can be combined with any of the first to fourth options, wherein the data sequence is a binary data sequence, and sub-bit periods or patterns of logic values ​​different from the current logic value of the binary data sequence (e.g., second binary state and / or idle state (tri-state current)) can be inserted into the drive signal to achieve an upper or lower target level. This measure ensures that the radiated output can be actively corrected to ensure an appropriate level at the sampling time.

[0032] The sixth option, which can be combined with any of the first to fourth options, allows selection of either an upper or lower target level, such that the rise time of the radiated output from the lower to the upper radiated output level is equal to the fall time of the radiated output from the upper to the lower radiated output level. Therefore, when the logic values ​​are uniformly distributed in the binary data sequence, a steady-state radiated output level can be ensured.

[0033] According to the seventh option of any of the first to fourth aspects, which can be combined with any of the first to sixth options, the drive signal can be modulated to produce more than two radiated output levels, wherein the transition in the trellis of the multi-level drive mode is adapted to the rise and fall curves of the radiation source. Therefore, a trellis-based multi-level transmission system can be provided, wherein the bit rate is adjusted based on transmission quality.

[0034] Note that the above-described device may be implemented based on an arrangement of discrete hardware circuits, integrated chips, or chip modules with discrete hardware components, or based on a signal processing device or chip controlled by software routines or programs stored in memory, written on a computer-readable medium, or downloaded from a network (such as the Internet).

[0035] It should be understood that transmitters, optical communication systems, methods, and computer program products may have similar and / or identical preferred embodiments.

[0036] It should be understood that the preferred embodiments of the present invention may also be any combination of the above embodiments.

[0037] These and other aspects of the invention will become clear and explained with reference to the embodiments described below. Attached Figure Description

[0038] In the following figures: Figure 1 A block diagram of an optical communication system according to various embodiments is shown schematically; Figure 2 Flowcharts of the driving process according to various embodiments are shown; Figure 3 A flowchart illustrating the receiver feedback process according to various embodiments is shown; Figure 4 The waveforms of the optical response at different switching speeds according to various embodiments are schematically shown. Figure 5 The diagrams schematically illustrate the binary data sequence, light source current, and corresponding waveforms of the resulting light output according to various embodiments. Figure 6 A waveform diagram illustrating switching between logical states of optical output according to various embodiments is shown schematically. Figure 7 The waveforms of the light source current and the resulting light output are schematically shown according to various embodiments; Figure 8 A circuit diagram of a driver with optical output feedback according to one embodiment is shown schematically. Figure 9 The diagram schematically illustrates the waveforms of the light source current and the resulting light output in a high-rate switching mode according to one embodiment; and Figure 10 The diagram schematically illustrates the rise and fall curves of the light output of a grid-coded modulation scheme and a light source according to one embodiment. Detailed Implementation

[0039] Various embodiments of the invention will now be described based on optical wireless lighting and communication (LiFi) systems with multiple transceiver access devices. While the invention is particularly advantageous in the context of lighting systems, it is not limited thereto and can also be used in optical wireless communication systems not integrated within lighting systems or not integrated within fiber-optic-based optical communication systems.

[0040] In the following text, a light source can be understood as a radiation source that produces visible or invisible light for communication purposes (i.e., including infrared (IR) or ultraviolet (UV) light sources). The light source can be contained within a luminaire, such as an incandescent lamp, fluorescent lamp, or other discharge lamp that is recessed or surface-mounted. The luminaire can also have non-traditional types, such as fiber optic components in one location and an optical fiber core or "light guide" in another. These concepts can also be used for peer-to-peer communication between smartphones or Internet of Things (IoT) devices.

[0041] It is also important to note that when using optical wireless communication based on the invisible portion of the spectrum (such as infrared and / or ultraviolet), the system can be completely isolated from any lighting system. In this scenario, the optical wireless communication system can be used primarily to provide communication, and a separate transceiver node can be used within the optical wireless communication system. Alternatively, such an optical wireless communication system can supplement additional functionality and thus be integrated into other application devices that benefit from this communication capability; such as personal computers, personal digital assistants, tablets, mobile phones, televisions, etc.

[0042] Traditional light source luminaires are being rapidly replaced by light-emitting diode (LED) or laser-based lighting solutions. In LiFi systems, more advanced LED or laser-based luminaires can act as LiFi communication hubs, adding LiFi connectivity to the lighting infrastructure. The basic idea is that the lighting infrastructure is positioned in such a way that it provides a line of sight from the luminaire to the locations where people tend to live. As a result, the lighting infrastructure is also well-positioned to provide optical wireless communication that also requires a line of sight.

[0043] According to various embodiments, adaptive or accelerated on / off keying (OOK) schemes are provided for such LiFi systems or other optical communication systems to accelerate transmission rates while keeping the driver or drive circuitry simple and power efficient.

[0044] Figure 1 A block diagram of an optical communication system according to various embodiments is shown schematically.

[0045] Note—throughout this disclosure—only those structural elements and functions that aid in understanding the embodiments are shown. For the sake of brevity, other structural elements and functions have been omitted.

[0046] Figure 1 The optical communication system can correspond to the communication link of a LiFi network, which includes a transmitter (optical transmitter) 10 (e.g., an access point (AP) of a luminaire with a lighting system) connected to a receiver (photodetector) 20 via an optical channel (e.g., an optical free-space link). A corresponding light output 100 (e.g., a light beam) generated by a light source (LS) 12 of the transmitter 10 is received by a photodetector (PD) 22 of the receiver 20. The light source 12 may include a radiation emitting element (e.g., an LED or a laser diode), and the photodetector 22 may include a radiation detection element.

[0047] The transmitter 10 depicted herein includes an encoder (ENC) 16 for encoding input data DI received via interface circuitry (not shown) to obtain a binary data sequence data, which here consists of a sequence of binary values ​​"0" and "1" according to a selected binary encoding scheme. As illustrated by the dashed outline of the encoder ENC, this encoder may also be external to the transmitter. The binary data sequence data is provided to an adaptive OOK controller (A-OOK) 18, which controls a modulator circuitry (e.g., a switching circuitry) 14 to generate a drive signal DS (e.g., drive current or voltage) according to a proposed enhanced drive scheme and provides it to a light source 12 to generate an optical output 100 using accelerated OOK or other keying-based modulation schemes. More specifically, the adaptive OOK controller 18 may be configured to determine at least one of a switching time or rate (e.g., bit rate BR) for the drive signal DS. In one embodiment, the adaptive OOK controller 18 provides an amplitude and timing-adjusted data sequence data' to the modulator circuitry 14, where the modulator circuitry is essentially a radiation source driver. Alternatively, the adaptive OOK controller can also transmit the data sequence as well as the maximum and minimum permissible light output level LL or output light range of the light source 12, thereby offloading more timing-critical operations to the modulator.

[0048] Simultaneously, the maximum and minimum optical output levels, or the reduced radiation output range, are based on transmission quality information provided to controller 18. As mentioned above, transmission quality feedback can be determined solely within transmitter 10 and can be derived, for example, from the voltage on the semiconductor radiation source LS (in... Figure 1 In the middle, quality feedback Q1). Alternatively and / or additionally, it can be based on quality feedback Q2 fed back from receiver 20.

[0049] For example, assuming an implementation using LL and BR, as transmission quality increases, the drive signal DS and therefore the drive current can switch at a higher rate (increased bit rate) to reduce the distance between the maximum and minimum optical output levels, or the maximum and minimum optical output levels can be set closer together so that the drive signal switches at a higher rate.

[0050] Furthermore, the light source 12 can be configured to provide a feedback signal Q1 to the modulator circuit 14, which indicates the light output level or an attribute or parameter related to the light level. The modulator circuit 14 uses the feedback signal Q1, along with the switching time / rate and / or maximum / minimum light level or light range, to generate or control a drive signal DS and apply it to the light source 12. Based on the feedback signal Q1 and control information (BR, LL), the modulator circuit 14 adjusts the drive signal DS to control the range and / or level of the drive current of the light source 12 according to a proposed acceleration keying-based modulation scheme (e.g., accelerated OOK), and thereby controls the light output 100.

[0051] In one example, modulator circuit 14 (i.e., modulator driver) can act as a switching device for controlling drive signal DS to switch the drive current (or voltage across light source 12) through light source 12 (e.g., an optical transmitter) between several discrete values ​​(e.g., 2 or 4 discrete values).

[0052] At receiver 20, the output signal of photodetector 22 can be provided to demodulator circuit (DEM) 24, where the output signal is demodulated by detecting or identifying the optical output level to obtain a binary data sequence. This binary data sequence can then be decoded in decoder circuit (DEC) 26 to obtain output data that should correspond to the original input data (i.e., the original binary data sequence) DI provided to transmitter 10. Error detection circuit (ERR) 28 can then check the output data based on an error detection scheme (e.g., parity check, cyclic redundancy check (CRC), error correction coding, etc.) to determine the transmission quality (e.g., signal-to-noise ratio) of the optical transmission. The check result can optionally be fed back from receiver 20 to transmitter 10 via optical or other wireless channels as transmission quality information used by adaptive OOK controller 18.

[0053] In one example, the control software may run on the central processing unit (CPU) provided in the adaptive OOK controller 18 and / or receiver 20 to provide the controller and receiver functions discussed herein.

[0054] As a semiconductor device (such as an LED or laser), the physical properties of the light source 12 make the light output 100 a low-pass filtered variant of the drive current. More specifically, the junction of the semiconductor light source is a capacitor, and the discharge of this capacitor by means of electron-hole pairs recombine into photons is a nonlinear function of charge. In particular, when the capacitor is in a low-charge state, not many photons are generated, making the discharge increasingly slow. If the symbol rate (Nyquist) bandwidth of the drive signal DS is greater than the 3 dB bandwidth of the light source, ISI will occur without further measures. However, the switching time / rate (e.g., bit rate (BR)) of the light source 12 corresponding to the time resolution of the on-off switching can be selected by the adaptive OOK controller 18 to be faster than the symbol rate or to occur at other (later or earlier) moments outside of the bit transition. As a result, it can be facilitated that the light output reaches the target light level value at the sampling time.

[0055] According to various embodiments, it is intentional that the on-state of the drive current used to turn on the light source is greater than the drive current that results in the desired steady-state light output. Additionally, the off-state of the drive current used to turn off the light source 12 may not be zero current (i.e., the drive current is disconnected from the current source).

[0056] In one example, during the off-state of the drive current, light source 12 can be short-circuited to achieve faster discharge of light source 12. Therefore, during the off-state, an "above-demand" on-state current and a "short-circuited" light source 12 can be combined to achieve accelerated OOK.

[0057] In another example, during the light source's off-time, an intentional negative drive current can be applied as a turn-off current ("active sweep-out"). This can be achieved, for example, by a transistor having a source-drain or collector-emitter connection across the light source. This sweep-out current can be applied to reach the light output level corresponding to logic zero more quickly. However, for multiple consecutive zeros, a fully depleted junction may not be attractive. In this case, recovering to the charged junction state (requiring a logic one) would be very long. Therefore, measures are taken to ensure that after a rapid sweep-out during the first zero-point period, the junction of light source 12 remains lightly charged during subsequent zero-point periods. This can be achieved by injecting short pulses of turn-on current. Figure 2 It shows that according to what can be seen Figure 1 The flowcharts show the driving processes of various embodiments implemented in the adaptive OOK controller 18.

[0058] In the first step S201, the transmission quality of the optical channel (e.g., quality of service (QoS), signal-to-noise ratio (SNR), bit error rate (BER), symbol error rate (SER), etc.) is determined, for example, based on measurements from the transmitter side or feedback information from the receiver side. Then, in step S202, the range of bit rate and / or optical output level (or maximum and / or minimum optical output level) used to drive the transmitter light source is determined based on the determined transmission properties (e.g., gradually increasing or decreasing) and is used to modulate the transmitter light source.

[0059] In step S203, the current binary data sequence or pattern used to modulate the transmitter light source is monitored. If, in subsequent step S204, it is determined that the monitored data sequence includes multiple consecutive levels of the same binary value (e.g., a sequence of two or more "0" values ​​or a sequence of two or more "1" values), the process continues to step S205, where the modulation or switching bit rate of the transmitter light source is increased by a predetermined amount, and a predetermined sub-bit pattern is introduced during the multiple consecutive levels of the same binary value to maintain the optical output level at a determined maximum or minimum optical output. Otherwise, if the current binary data sequence or pattern does not include multiple consecutive levels of the same binary value, the process branches to step S206, where the transmitter light source is driven at the predetermined bit rate determined in step S202.

[0060] Finally, the process jumps back to the initial step S201 and continues as long as data is available for transmission.

[0061] In a real system, steps S203 to S206 can run at very high speeds (e.g., hundreds of megabits per second) and can therefore be implemented in hardware. The bit rate adaptation in steps S201 and S202 can be performed slowly (e.g., at a rate of approximately one or several seconds) and in response to changes in the channel (e.g., by movement of one or more client devices). Furthermore, this bit rate adaptation may require protocol overhead to align the transmitter and receiver sides, which can be done in (embedded) software and can be performed every few (hundreds) milliseconds. Therefore, the level-switching loop and adaptive bit rate control can be separated into different processing flows.

[0062] In another embodiment, Figure 2The process can be improved to implement a three-state driver with a first state, a second state, and a third (idle) state. During the first state, a first current (e.g., a positive current) is provided to pull up the charge in the transmitter source. During the second state, a second current (e.g., a negative current) is provided to sweep out the charge from the transmitter source. During the third (idle) state, no current flows. The idle state can facilitate the emission of multiple consecutive logic zeros (or logic ones) during which the junction of the transmitter source tends to discharge slowly due to (photon and non-radiative) hole-electron recombination, making it possible that only a short pulse of positive current is needed during the sub-bit interval. Rapid alternation of this positive and negative current pulses can lead to excessive energy waste, while alternation between positive current and the idle state is more energy-efficient. Therefore, this three-state operation may involve a high positive current used to pull up the charge in the transmitter source during transitions to higher optical output levels. The negative sweep-out current can be used to accelerate transitions to lower optical output levels. The idle (zero current) mode can be used when small changes in the optical output level are not expected. The idle state can be alternated with short positive pulses to (re)charge the junction of the transmitter light source. Figure 3 A flowchart of a receiver feedback process according to various embodiments is shown, in which transmission quality information is fed back from the receiver side to the transmitter side.

[0063] In step S301, the received data sequence is evaluated (e.g., after decoding or demodulation) to determine the transmission quality of the optical channel (e.g., signal-to-noise ratio (SNR), bit error rate (BER), symbol error rate (SER), etc.). Then, in step S302, it is checked whether a predetermined threshold for at least one transmission quality parameter (e.g., signal-to-noise ratio (SNR), bit error rate (BER), symbol error rate (SER), etc.) has been exceeded. If so, the process continues to step S303, in which a request to reduce the bit rate is fed back to the receiver side, for example via the optical channel or RF channel (NFC, Bluetooth, WiFi, etc.). Otherwise, if the threshold has not been exceeded, the process jumps back to step S301 and continues as long as transmission is in progress.

[0064] Alternatively, in step S302, it can be checked whether the threshold has been exceeded, and in step S303, the feedback information notified by signal can be a request to maintain the bit rate, while a request to reduce the bit rate can be notified by signal in other ways.

[0065] Therefore, according to various embodiments, the full range of optical levels that can be used for OOK or other keying modulations is limited to a reduced swing or range of optical output levels determined by the minimum and / or maximum optical output levels, so that the optical output levels can still be distinguished on the receiver side, although the eye diagram opening is smaller due to the filtering characteristics of the transmitter light source at the selected higher bit rate.

[0066] Furthermore, in cases where multiple consecutive levels of the same binary value result in 0-to-0 and 1-to-1 bit transitions, a "sub-bit" switching ("flickering") at an increased bit rate can be introduced to maintain the optical level (instead of letting it drop or rise further). This reduces ISI. While this measure can reduce the distance between the maximum and minimum optical output levels to some extent, it offers the advantage of increasing the ISI-free bit rate (i.e., increasing transmission speed) when the transmission quality is high enough.

[0067] In one embodiment, the modulator (e.g., Figure 1 The controller 18 and / or modulator circuit 14 in the middle utilize the transmitter light source (e.g., by switching the light source in such a way) Figure 1 The known property of the low-pass characteristics of the light source 12 in the light source is that at a given sampling time, the light output signal reaches one of a plurality of discrete light output levels (e.g., the maximum and minimum light output values ​​determined in the case of a two-level system).

[0068] In one example, the modulator can use timing advance or timing delay during switching operations to target a desired optical output level. Within the interval of transmitted symbols (e.g., logic "0"), the modulator can insert sub-bit periods or patterns (e.g., current on / off) with logic values ​​different from the logic value of the current symbol to achieve the desired optical output level of the transmitter source at the sampling time.

[0069] In another embodiment targeting optimized optical output levels, the modulator (e.g., Figure 1 The controller 18 and / or modulator circuit 14) can achieve the desired low optical output level ϕ0 and the desired high optical output level ϕ1, such that the optical output levels ϕ0 and ϕ1 can be detected by the receiver as representing logic levels "0" and "1" respectively, the optical output level ϕ0 with logic value "0" is strictly positive and higher than the steady-state optical output level when the drive current is continuously switched to its minimum value, and the optical output level ϕ1 with logic value "1" is less than the steady-state optical output level corresponding to the continuous switching of the drive current.

[0070] In one example, the selection of output light ϕ0 and ϕ1 can be optimized such that the rise time from ϕ0 to ϕ1 is equal to the fall time from ϕ1 to ϕ0 (where the turn-off current begins at f1). For example, for two consecutive logic values ​​"0", the modulator can ensure that the light output level of the transmitter source is equal to ϕ0 at the corresponding two consecutive sampling times t1 and t2, for example, by applying turn-on and turn-off currents during a segment (sub-bit period) between sampling times t1 and t2. Similarly, for two consecutive logic values ​​"1", the modulator can ensure that the light output level of the transmitter source is equal to ϕ1 at the corresponding two consecutive sampling times t1 and t2, for example, by applying turn-on and turn-off currents during a segment (sub-bit period) between sampling times t1 and t2.

[0071] Figure 4 The diagram schematically illustrates the waveforms of the light output (photoresponse) of the transmitter light source as a function of time for normal OOK and accelerated OOK (A-OOK) at different switching speeds, according to various embodiments.

[0072] Different rise and fall characteristics with different rise and fall times were observed in response to the drive current supplied to the transmitter light source (e.g., LED) when the switch is turned on and off. Figure 4 The upper waveform shows a typical OOK waveform at a rate below the transmitter's light source bandwidth, where sufficient time is available for the light source current, and thus the optical output level reaches a zero state (i.e., zero drive current (ZC) or turn-off current), and reaches full optical output level when a full drive current (FC) or turn-on current is applied.

[0073] in addition, Figure 4 The lower waveform shows the proposed A-OOK driven output optical waveform, which limits the level swing or range in the optical output and allows for faster modulation. Because the increased bit rate exceeds the bandwidth of the transmitter light source, the available time of the light source current, and therefore the optical output level, is no longer sufficient to reach zero optical output level and full optical output level, respectively. As a result, the optical output varies between a minimum optical output level ϕ0 (greater than zero) and a maximum optical output level ϕ1 (lower than the maximum level of the OOK waveform). Figure 4 (as shown in the diagram above), which allows for the acquisition of the offset level OS of the optical output. Therefore, the A-OOK driving scheme (e.g., switching time / rate, maximum and / or minimum optical output level) is controlled based on the transmission quality of the optical channel to allow the logic output level to be distinguished at the receiver at an increased bit rate.

[0074] Figure 5 The diagram schematically illustrates the waveforms of binary data sequences, light source current, and light output according to various embodiments.

[0075] The waveform above indicates a binary data sequence (OOK-BD) with normal OOK modulation used to drive the transmitter light source as a function of time.

[0076] Furthermore, the intermediate waveform shows the waveform generated by the modulator (e.g., according to the proposed A-OOK pattern) based on the proposed A-OOK pattern. Figure 1 The drive current (I) supplied to the transmitter light source after processing by the controller 18 and / or modulator circuit 14 LS () as a function of time.

[0077] Finally, the waveform below shows the light output (L) of the transmitter light source. out (The curve) and binary drive current, as a function of time caused by the switching cycle (bit rate), differ from typical OOK symbol timing. The symbol value is reached, and the symbol value is indicated by a black circle. Here, levels ϕ0 and ϕ1 represent the target optical output values ​​of logic zero and logic one, respectively, which will be reached at the sampling time. However, in any other instance, possible deviations outside the sampling time are limited to ensure that the correct optical output level is reached in time at the sampling time.

[0078] As from Figure 5 As can be seen from the middle and lower waveforms, the 1-to-1 conversion of the binary data sequence is modified by shortening the second time period of the on-current to maintain the maximum optical output level ϕ1. Additionally, the 0-to-0 conversion of the binary data sequence is modified by adding a shortened (i.e., sub-bit) time period of the on-current to maintain the minimum optical output level ϕ0.

[0079] In A-OOK drive mode, the on-off switching of the transmitter light source drive current is selected such that a particularly intentionally chosen maximum and / or minimum optical output level is reached during the sampling time. In one example, the switching pattern of the drive current does not necessarily follow the binary pattern of the binary data sequence, especially during many consecutive bits of the same value.

[0080] In the first example of the A-OOK drive mode, the maximum and / or minimum optical output levels can be achieved or maintained by changing the timing of the drive current on or off to the sub-bit timing.

[0081] In the second example of the A-OOK drive mode, for example during the “011” bit sequence, the maximum and / or minimum optical output level of an intentionally selected method (delay method) can be achieved or maintained by keeping the current on for a sub-bit period after the first binary value “1”, which is a predetermined share of the bit duration, but then turning off the drive current before the end of the transmission interval of the second binary value “1”.

[0082] In the third example of the A-OOK drive mode, for example during the “001” bit sequence, by turning on the drive current before the transmission interval of the binary value “1” begins, an intentionally selected maximum and / or minimum optical output level can be reached or maintained to ensure that two consecutive binary values ​​“0” are received at the same minimum optical output level (time-advancement method).

[0083] In the fourth example of the A-OOK drive mode, for example, during the operation of multiple consecutive bits of the same binary value, the intentionally selected maximum and / or minimum optical output level can be achieved or maintained by inserting short on or off cycles of the drive current respectively.

[0084] In the fifth example of the A-OOK drive mode, for example during the “000” bit sequence, the intentionally selected maximum and / or minimum optical output level can be achieved or maintained by inserting short sub-bit on-time cycles of the drive current to maintain the minimum optical output level.

[0085] Figure 6 A waveform diagram illustrating the switching between logical states of light output as a function of time, according to various embodiments, is shown schematically.

[0086] More specifically, for the conversion of "01" and "10" in binary data sequences respectively, in Figure 6 The diagram shows the switching between the "logic one" optical output level ϕ1 and the "logic zero" optical output level ϕ0, which follows the rising curve ϕ. ON (t) and the descending curve ϕ OFF (t). For example, from... Figure 6 As can be seen from this, the switching period between the on and off states of the drive current can be determined, depending on the bit period or duration T. b Maintain the desired minimum and maximum optical output levels ϕ0 and ϕ1 and their deviations during the "00" and "11" transitions. If the bit duration T... b If the (i.e., the reciprocal of the bit rate) exceeds the time interval required for the transmitter light source to reach ϕ1 from ϕ0 and the time interval required for the transmitter light source to reach ϕ0 from ϕ1, then ISI-free transmission of binary data sequences from the transmitter to the receiver can be achieved.

[0087] The off state of the drive current can also be a state that short-circuits the transmitter light source. Similarly, in an embodiment, the on state of the drive current can be a predetermined voltage level required to power the transmitter light source for a predetermined rise time.

[0088] Figure 7 The diagram schematically illustrates the waveforms of the light source current as a function of time and the resulting light output according to various embodiments.

[0089] In this example, the lower limit ϕ is inverted when the light output is monitored. LRL and the inverted upper limit ϕ URL It is used as direct feedback. For example, by Figure 7 The triangle “Δ” in the diagram indicates that whenever the modulator (e.g., Figure 1 The controller 18 and / or modulator circuit 14 in the middle determine the rising curve ϕ ON (t) or the descending curve ϕ OFF The determined light output level ϕ (t) reaches the inversion lower limit ϕ respectively. LRL Or reverse the upper limit ϕ URL At that time, the drive current is turned on or off respectively to obtain the desired maximum optical output level ϕ1 or minimum optical output level ϕ0 at the sampling time, as shown by... Figure 7 The black circle indicates this.

[0090] Therefore, the upper and lower inversion limits are threshold levels. When these threshold levels are exceeded, the driver is triggered or switched to switch the drive current to another state (e.g., the opposite logic state or an idle state) so as not to exceed the inversion limits. The inversion limits can be set based on the rising and falling curves to obtain the desired maximum optical output level ϕ1 or minimum optical output level ϕ0 at the sampling time. More specifically, when the drive current is on and the rising optical output level reaches the upper inversion limit ϕ... URL When the drive current is turned off ( Figure 7 (The upper triangle in the waveform). On the other hand, when the drive current is off and the falling optical output level reaches the inversion lower limit ϕ. LRL At that time, the drive current is turned on.

[0091] Figure 8 A circuit diagram of a driver with optical output feedback according to an embodiment is shown schematically.

[0092] exist Figure 8 In an exemplary circuit diagram, the voltage V across the transmitter light source (e.g., LED) 84 mod It can be used as an indicator of the junction charge of light source 84, and thus as an indicator of the light output level. Voltage V mod The voltage level can be used as direct feedback for the switching transistor 86 of the modulator, for example, to control the switching of the drive current accordingly. To achieve this, the voltage V... mod The input is applied to the first input of comparator 88. The second input of comparator 88 serves as a reference, with the reference value (which corresponds to, for example...) Figure 7 upper limit of reflection ϕ URL / Reflection lower limit ϕ LRLA signal is applied to the reference terminal. When the light source 84 is considered as a capacitor in parallel with the resistor, the switching transistor 86 charges and discharges the light source 84 to the level corresponding to the bias input (e.g., ...). Figure 4 The level of the offset value OS in the middle, which has or does not have the amplitude of the binary optical output level.

[0093] The reference value can be generated by level shifter 82 (e.g., a summing operational amplifier with a predetermined amplification factor), and the binary data sequence OOK-BD to be transmitted by light source 84 and the bias B (e.g., the offset OS relative to the minimum light output level of light source 84) are applied to the two summing inputs of level shifter 82.

[0094] The output voltage of level shifter 82 can therefore be set to correspond to the upper reflection limit ϕ. URL / Reflection lower limit ϕ LRL As a result, the switching transistor 86 can be controlled by the comparator 88 to achieve the upper reference value (e.g., the inverted upper limit ϕ) at the rising light output level. URL When the transmitter light source drive current is turned off, and the falling light output level reaches the lower reference value (e.g., the inverted lower limit ϕ), the light output level is reduced. LRL When the drive current is turned on... In another embodiment, the drive control of the transmitter light source is configured to maintain the optical output level during two consecutive bits of the same binary value. This can be achieved by selecting an appropriate duty cycle or duty rate for the on and off times of the drive current between the sampling moments of the two consecutive bits.

[0095] In one example, the duty cycle for this level is not exactly 50%. If the duty cycle were exactly 50%, the lower or minimum optical output level ϕ0 would drift upwards, and the higher or maximum optical output level ϕ1 would drift downwards. Eventually, after a very long sequence, the higher and lower optical output levels would reach a medium optical level.

[0096] In another embodiment, a memory (e.g., a lookup table) can be used to store a predetermined number L of previous symbol values ​​(i.e., n-1, n-2, ... n-1) of the binary data sequence, where L can be greater than or equal to 1. Optionally, the memory effect for storing previous values ​​can be achieved by connecting an inductor to the transmitter light source with a wire.

[0097] Values ​​stored in a lookup table or other type of memory can be used to determine the high-rate (i.e., sub-bit) switching mode of the drive current to be supplied to the transmitter light source during the bit duration, for example, as indicated in the exemplary table below: 00 0 000011 01 0 100000 10 0 000111 11 0 000000 00 1 111111 01 1 111100 10 1 011111 11 1 111100 .

[0098] In the example in the table above, if two previous bit values ​​were already "00", then during the bit duration of the new bit value "0", the high-rate drive current switching mode "000011" is applied to the transmitter light source. That is, the drive current is turned off for the first two-thirds of the bit duration and then turned on for the second third of the bit duration.

[0099] On the other hand, if the two previous bit values ​​are already "01", then during the bit duration of the new bit value "0", the high-rate drive current switching mode "100000" is applied to the transmitter light source. That is, the drive current is turned on for the first one-sixth of the bit duration and then turned off for the remaining five-sixths of the bit duration.

[0100] In another scenario, if both previous bit values ​​were already "00", then during the bit duration of the new bit value "1", the high-rate drive current switching mode "111111" is applied to the transmitter light source. That is, the drive current is on for the entire bit duration.

[0101] Of course, one or more preceding bits can be used to determine the high-rate switching mode of the drive current. Furthermore, shorter or longer switching modes can be used (i.e., less than or more than six sub-cycles of bit duration).

[0102] Figure 9 The light source drive current (I) with a high-rate switching mode according to the table above is schematically shown. LS ) and the resulting light output as a function of time (L out The waveform diagram.

[0103] Therefore, the driving current through the transmitter light source (e.g., LED) is controlled by the modulator (e.g., ... Figure 1 The controller 18 and / or modulator circuit 14) are controlled using the lookup table shown above.

[0104] according to Figure 9 The binary data sequence "11000101" will be signaled by the transmitter light source. Based on the table above, the following eight high-rate switching modes of the drive current need to be provided to the transmitter light source to signal the aforementioned binary data sequence, while ensuring that the optical output level remains at the minimum response maximum optical output level ϕ0 and ϕ1 at the sampling time (i.e., Figure 9 (The black circle in the middle) "111111" is used for the first bit value "1" that has the previous sequence "00"; "111100" is used for the second bit value "1" that has the previous sequence "01"; "000000" is used for the third bit value "0" that has the previous sequence "11"; “000111” is used for the fourth bit value “0” that has the previous sequence “10”; “000011” is used for the fifth bit value “0” that has a previous sequence of “00”; "111111" is used for the sixth bit value "1" that has the previous sequence "00"; "100000" is used for the seventh bit value "0" that follows the previous sequence "01"; and "011111" is used for the eighth bit value "1" which has the previous sequence "10".

[0105] As from Figure 9 As can be seen, a high-speed switching mode has been selected to meet the inversion lower limit ϕ of the obtained optical output level. LRL and the inverted upper limit ϕ URL .

[0106] As described earlier, depending on the transmission quality of the optical channel (e.g., noise level), for binary values ​​"0" and "1", the receiver needs a certain distance between the lower and higher optical levels received. Therefore, at poor signal-to-noise ratios (SNR), the communication bit rate should be reduced to ensure that each bit has sufficient energy for reliable communication.

[0107] As an example, in longer communication ranges or distances where the received signal is weak, the bit rate used for communication can be selected to be less than 3 dB bandwidth of the transmitter light source. As a result, the transmitter light source has sufficient time for its optical output to follow the switching mode of the drive current and reaches a zero / one level for a portion of the bit duration, making the received signal waveform resemble the waveform of the switching current and providing the highest achievable energy per bit.

[0108] Otherwise, in short communication ranges or distances where the received signal is strong enough to overcome the noise level, the swing or range between the one and zero optical output levels (i.e., eye diagram opening) can be reduced. Within very short communication ranges or distances, a major limitation is how quickly the transmitter's light output can increase from 0 to 1 or decrease from 1 to 0, allowing the one and zero optical output levels to be spaced just far enough to overcome the (low) noise level.

[0109] Therefore, a rate-adaptive transmission system can be provided, which is configured to maximize the communication bit rate based on the conditions or characteristics of the optical channel.

[0110] Therefore, as Figure 1 One of the options shown is the modulator (e.g., Figure 1The controller 18 and / or modulator circuit 14 can use the feedback channel from receiver 20 to transmitter 10 to control the bit rate. If the selected bit rate is too high, the modulator can reduce the bit rate to reduce the bandwidth and thus improve the transmission quality (e.g., SNR per bit). In addition, because the on or off state of the drive current is maintained for a longer duration, a larger swing or range of optical output level is achieved.

[0111] Otherwise, if the chosen bit rate is so small that sufficient transmission quality (e.g., SNR margin) is available, the modulator can increase the bit rate.

[0112] Therefore, reducing the bit rate (or increasing the bit duration) allows the waveforms and optical levels of binary values ​​"0" and "1" to deviate more, which increases the distance between the lower and higher levels of the optical output. Secondly, if the bit duration becomes longer, the noise bandwidth can be further reduced.

[0113] More specifically, at the sampling time, the distance between the two levels can be estimated as follows: Where T b It represents the duration of a bit, and the value of the parameter ϕ' depends on the history of the bit sequence (i.e., the bit values ​​before the current bit value).

[0114] In another embodiment, the operating point of the driving scheme can be automatically set. For the equal number of zeros and ones in the binary data sequence to be transmitted via the light source, the operating point of the average light output automatically converges to a predetermined center value.

[0115] For very high bit rates with a 50% duty cycle (e.g., symmetrical square waves, on / off keying), if the rise and fall times are different, the average optical output will not reach half of the maximum optical output level ϕ1. A balance emerges where the decrease in optical output during the off-phase of the drive current precisely balances the increase in optical output during the on-phase of the drive current.

[0116] A sufficient condition for stable equilibrium (operating point) is: 1) The rising curve ϕ of light output ON (t) monotonically increasing (i.e., ϕ') ON (t)>0) and the second derivative is negative (i.e., ϕ" ON (t)<0), and 2) The decrease curve of light output ϕ OFF (t) monotonically decreasing (i.e., ϕ') ON (t)<0) and the second derivative is positive (i.e., ϕ" OFF (t)>0).

[0117] Therefore, it can be assumed that if the bitstream is short-term balanced, that is, if it has approximately the same number of logic zeros and ones, the modulator will automatically set sufficient levels for stable operation of the proposed adaptive drive scheme (e.g., accelerated OOK).

[0118] For longer communication ranges or distances, transmission quality (e.g., SNR) is lower. In this case, the bit duration must be increased so that the optical output follows a rising and falling curve over a longer period. Eventually, it will reach a steady-state level. In other words, if transmission quality requires utilizing the full dynamic range of the LED light output, the adaptive drive system converges to a conventional OOK system.

[0119] Various options may be used for optimization.

[0120] In another embodiment, the upper limit of the radiated output is set to a maximum value that will not damage the LED. This can be considered, for example, that after a long period of high data levels (a long string of logic 1s), the associated optical power level does not excessively heat the LED, or similarly, the associated time-averaged data-driven drive current DS does not exceed the absolute maximum rated value of the LED current. Alternatively, when the data bitstream is encoded as a balanced stream (i.e., with an equal number of 1s and 0s) and balanced even within a shorter time window, then the short-term drive current DS is averaged over a short sequence in the data sequence and can be up to twice the absolute maximum current. In the case of such short-term balanced data, the upper limit of the radiated level can be up to twice the radiated level that would long-term limit the LED's light output.

[0121] In this embodiment, the lower radiation limit can be adapted as a trade-off between bit rate and achievable communication range. Setting the lower radiation limit only below the upper limit speeds up the system but makes it susceptible to noise and has a limited range. Setting the lower limit closer to zero makes the system's bit rate slower but more robust to noise and thus allows for a longer transmission range. In the two states where the upper limit is chosen to not damage the LED, the falling curve is likely to impose / define the shortest bit duration (and thus limit the bit rate). On the other hand, if the rising curve will impose / define the shortest bit duration (and thus limit the bit rate), it would be attractive to consider reducing both the upper and lower limits of the radiated output, thereby maintaining the distance between the upper and lower radiation limits but allowing for faster transitions.

[0122] Alternatively, when the goal is not the fastest possible transmission rate but a reasonable bit duration, this leads to the following design rule for the positive drive current. If the maximum non-destructive optical output level of the light source is known, and the maximum range is known (and therefore the required distance between high and low optical levels) and the required highest bit rate (and therefore the shortest bit duration), then the positive drive current needs to be just large enough to raise the optical level from low to high in the shortest bit duration.

[0123] Because of physical limitations (e.g., thermal limitations) on the maximum optical level, the downward current value can be designed for the most demanding range and bit rate combinations. Within this range, the distance between the received higher and lower optical levels must be sufficient to resist noise. This sets a lower limit (which should be at most equal to the upper limit minus the distance). The negative drive current should then be low enough (negatively strong enough) to achieve the minimum optical level within the shortest bit duration.

[0124] In another embodiment with simplified electronics, the negative current is implemented as a short circuit to the light source. This short circuit depletes the LED junction capacitance much faster than an open circuit, i.e., much faster than depleting the junction capacitance solely through recombination of internal holes (which could be slow if the LED is already partially discharged). In such a system, the receiver noise level and path loss require a minimum distance between the lower and higher light levels.

[0125] When considering power optimization at the bit rate, a lower optical level is chosen, and a higher optical level is also chosen by increasing the required distance, so that the fall time during short circuits decreases sufficiently quickly from the high optical level to the low optical level to support the bit rate. This process results in carefully selected upper and lower optical limits. The pull-up positive current is then chosen to be high enough that the rise time from the low optical level to the high optical level is as fast as or faster than the aforementioned fall time.

[0126] Throughout the preceding text, the focus has been on using on / off keying to achieve a two-level radiation output range. However, the method described herein is not necessarily limited to a two-level scheme. When the reduced radiation output range is sufficient to accommodate more levels, it is possible to implement a three-level approach in a similar manner, where on / off keying is used to implement three distinct levels, such as a first level "0" corresponding to the lower target level, a third level "2" corresponding to the upper target level, and a second level "1" corresponding to the midpoint between the lower and upper target levels.

[0127] When implementing a transmission scheme using a three-level radiated output, the reduced radiated output range needs to be sufficient for the receiving device to distinguish between the three levels at the receiver side. Consequently, the interval between the lower and upper target levels in the three-level scheme will therefore require a larger interval than that required in a two-level scheme. Furthermore, the timing of this multi-level encoding will be determined by the largest step size, which corresponds to the slowest level change. This is from a higher radiated level to a lower radiated level or from a lower radiated level to a higher radiated level. When using this three-level scheme, the transmission quality needs to capture the transmission quality of the slowest transition, allowing the adaptive OOK modulator to set the reduced radiated output range in a manner that will allow for correct reception. Moreover, since the reduced radiated range needs to accommodate all levels, and therefore will generally require a larger output range, the amount of space for timing variations will be reduced.

[0128] In addition to using multiple optical output levels, implementing multiple (e.g., three) drive current levels can also be beneficial. This can also be advantageous if only two optical levels are used. In a simple example, 1) a pull-up current is used to support the transition from low to high optical level, 2) a pull-down current is used to support the transition from high to low optical level, and 3) an idle current (with the light source open) is used to maintain the optical level transition from 1 to 1 or maintain a low optical level (0 to 0). For the third case, short pulses of current (during the sub-bit interval) are needed to maintain the charge level when the light source loses holes and electrons due to photon recombination. This three-level driver with an idle state consumes less power and results in less EMC radiation.

[0129] In yet another embodiment, a different multilevel system is proposed. This embodiment can prevent limitation on the optical output range. While it avoids ISI, it limits the difference between the two binary levels (0 and 1) of the optical output. Further separation of the optical output levels provided by the multilevel system during the duration of the same binary value (one or zero) in the input data sequence can be used to improve the “distance” between consecutive sampled values.

[0130] Figure 10 The multilevel grid-coded modulation scheme is illustrated schematically. Figure 10 (left side) and the light output rise and fall curves ( Figure 10 (the right part), as the fastest possible level transition. Figure 10 It reveals how the additional reachability provided by multilevel systems can be advantageously used in a controlled manner.

[0131] exist Figure 10 In the example, the four-level output signal of the transmitter light source can be created by appropriately switching the drive current of the light source on and off. Modulator (e.g., Figure 1The controller 18 and / or modulator circuit 14) are designed for four levels, but due to the rise and fall curves ϕ ON (t), ϕ OFF (t) Due to the finite turn-on and turn-off speeds, not all transitions are feasible. This results in a “grid” encoding of the signal. Compared to conventional grid error correction codes, the transitions in the grid selected for the proposed multi-level drive mode are made to conform to the low-pass constraints of the transmitter light source, i.e., to ensure that any transition used matches the rise and fall curves of the transmitter light source.

[0132] More specifically, depending on the transmission quality (e.g., SNR) of the determined optical channel, a specific distance between optical output levels must be maintained to ensure a sufficiently low bit error rate (e.g., BER) at the receiver.

[0133] For example, the distance between the four target optical output signal levels can be equidistant to provide the highest robustness against noise that might affect the optical output level detected at the receiver. The distance between two points can be chosen to ensure sufficient transmission quality (e.g., BER). If the quality is too low, a larger distance may be needed between the target optical output levels, which can be achieved by reducing the bit rate. The total set of target levels can be increased so that the lowest target optical output level is sufficiently above zero to distinguish it from noise levels, where the transmitter response is fast (similar to the two-level scheme described above, which should be sufficiently above the lowest achievable optical output level).

[0134] exist Figure 10 In the example, trellis-coded modulation is used, where the transition is designed based on the maximum possible step size allowed by the rise time of the low-pass characteristics of the transmitter light source, taking into account the limitation of the (amplified) drive current. Trellis-coded modulation (e.g., proposed by Ungerboeck at [https: / / en.wikipedia.org / wiki / Trellis_modulation]) was originally intended for telephone lines with hard bandwidth limitations. In contrast, semiconductor light sources (e.g., LEDs) have a relatively flat roll-off above their 3 dB bandwidth, which can be described as a first-order roll-off. Therefore, it is recommended to push the bit rate above the 3 dB bandwidth of the transmitter light source. Second, semiconductor light sources (e.g., LEDs) exhibit nonlinearity (e.g., in the form of different rise and fall times) and bandwidth that depends on the signal level. The selection of the output light levels of a multilevel system and the paths between these output light levels is recommended to be designed based on the response speed of the transmitter light source and what can be achieved by switching the drive current, which is greater than the drive current required to reach the steady state of the transmitter light source, or greater than the drive current that would be suitable if the transmitter light source is modulated primarily (or exclusively) within its 3 dB bandwidth.

[0135] At very high bit rates, it is clear that the transmitter light source is driven as if it were an integral channel, i.e., as if a (non-linear) capacitor were being charged and discharged.

[0136] In one example, the rise and fall curves of the transmitter light source can be used to determine the relationships between the first and third levels (1 to 3, the maximum rise from the slowest region) and the fourth and second levels (4 to 2, the maximum fall from the highest level), respectively. The bit rate, and therefore the duration of each symbol, can be set to allow only these output light levels for the optical output signal.

[0137] In another example, the set of all levels can be boosted to ensure that the rise / fall times of these 1 to 3 and 4 to 2 become the same.

[0138] More generally, the levels of a multi-level system can be chosen such that a two-step transition imposes at least substantially the same constraint on the time interval required between two symbol sampling moments as a one-step transition. In this example, a two-step transition from 1 to 3 and from 4 to 2 is more restrictive than a one-step transition from 3 to 4 and from 2 to 1. However, in another scenario, for example, if the communication range is relatively long, causing the system to attempt the maximum possible swing and thus relatively deep into the tail of the descent curve, the latter two (one-step) transitions may become more restrictive and may determine the maximum possible bit rate. A grid track can be followed by using only two current levels. The drive current can be used throughout the entire bit cycle to allow for maximum step sizes, such as from the first level to the third level or from the fourth level to the second level, while for smaller step sizes (e.g., from the second level to the third level or from the third level to the fourth level), some alternation or pattern of on and off currents can be used. This allows for efficient power driving of the transmitter light source, simple detection, and accelerated bit rates.

[0139] At any given time, the transmitter has only two options: transmit a first binary value (e.g., zero) or a second binary value (e.g., one). Within the physical constraints imposed by the transmitter's light source response and transmission quality, what steps can be taken depends on the current state (e.g., the current output light level). The proposed multilevel scheme acknowledges that the feasible step size within a one-bit time depends on the current state of the multilevel system. Three jumps (i.e., steps 1 to 4 and 4 to 1) may be impossible, but two steps 3 to 1 and 2 to 4 may also take more time than allowed because these are "deeper in the tail of the transmitter's light source's turn-off and turn-on curves." However, within a faster range of the transmitter's light source's rise and fall curves, two jumps 1 to 3 and 4 to 2 may become feasible. By traversing the constrained path above the grid, the range of the transmitter's light source's output light values ​​can be optimally used based on the current light output state and transmission quality, and error correction coding can be implicitly applied.

[0140] exist Figure 10 In the examples, the transitions from 4 to 2, 3 to 4, and 2 to 1 impose somewhat similar size constraints on the bit duration, as indicated by the rising / falling curve plotted as a solid line by the transmitter light source. Nevertheless, if the transitions in one direction (such as, for example...) Figure 10 The only constraints (4 to 2 and 2 to 1) allow for further optimization of the optical level selection. As... Figure 10 The rising and falling curves allow all upward transitions within the grid to occur in a shorter time. Therefore, all optical levels can be boosted with some vertical shift. Then, the fall time from the fourth level to the second level is reduced, but upward optical levels can also change faster. The optimal setpoint is reached when the slowest upward transition takes the same amount of time as the slowest downward transition.

[0141] For a four-level equidistant arrangement, the two degrees of freedom are the time distance between sampling moments and the level distance between optical levels. The former can be determined by transmission quality (e.g., SNR and BER), while the latter can be set by minimizing the slowest transition time between optical levels. Alternatively, the second and third levels can be placed closer together, thus shifting the first and third levels separately from the second and fourth levels. As another option, the second and third levels can be combined into a common level.

[0142] The receiver's decoder (e.g., Figure 1 The decoding operation at decoder 26 can be based on binary clipping operation.

[0143] In one example, the decoder can place a limiter between two intermediate levels of the multi-level optical output. Then, it can be observed that for a bit value "1", the optical output is at the third or fourth level, while for a bit value "0", the optical output is at the first or second level.

[0144] In another example, decoding can be achieved using a single-tap memory to store the previous bit value. If the previous bit value was "1", the receiver needs to detect a second or fourth level to determine whether the bit value is "0" or "1", respectively. The limiting level for the new bit can be at (or close to) the third level. Therefore, if the new bit value is "1", the transmit level is the fourth level, and if the new bit value is "0", the transmit level is the second level.

[0145] Furthermore, if the previous bit value was "0", the limiter level of the new bit value can be (or close to) the second level. Therefore, if the new bit value is "1", the transmit level is the third level, and if the new bit value is "0", the transmit level is the first level.

[0146] In another example, in the first step, grid decoding at the receiver can be achieved by limiting all levels of the received optical signal. Figure 10 In the example, four levels are distinguished. The decoder (e.g.) Figure 1 An example of a decoder (26) could be the Viterbi decoder. It can make a final decision on the value of a received bit at a later stage after the bit has already been received, because future bits provide further information needed to determine the bit value.

[0147] In the above embodiments, the sampling time can be determined by tracking the intermediate level of the received light and searching for the time position with the maximum eye opening (i.e., the maximum distance between the lower and higher light levels). An oscillator (e.g., a voltage-controlled oscillator (VCO)) can be used to generate a clock signal that controls the sampling time of the incident light signal (e.g., at a time slightly earlier or later than the detected maximum eye opening).

[0148] As mentioned above Figure 10The above example presents a summary of two current-level grids; in higher-level implementations, grids with more than two points can be deployed. In such implementations, the transmitter follows a step-limited path. In each step, the light level rises or falls by only one level. There is an exception at the grid boundaries (higher / lower levels). Here, for one of the two possible bit values, a step size of zero is used, while for the other bit value, a step size of one or two is implemented (depending on whether the receiver can distinguish two adjacent (or higher) levels). The latter typically depends on the properties of the semiconductor radiation source. In contrast, in the example above with a 4-level grid, a step size of 0 or 2 (instead of 1) is used at the boundaries.

[0149] For example, when using 7 photoelectric levels, the following transitions are allowed: 2 to 1, 2 to 3, 3 to 2, 3 to 4, 4 to 3, 4 to 5, 5 to 4, 5 to 6, 6 to 5, 6 to 7, plus boundary step sizes of 1 to 1, 1 to 3, 7 to 7, 7 to 5. Alternatively, the latter set of boundary step sizes can be chosen as 1 to 1, 1 to 2, 7 to 7, 7 to 6.

[0150] In the latter case, the system must take precautions to prevent the bit duration from being less than the time required to complete the slowest transition in a complete set of possible transitions. Furthermore, during transitions that could be achieved more quickly, it may be necessary to control the drive current (e.g., by switching the current polarity or switching to idle, or delaying the start of the current) to ensure it does not exceed the target.

[0151] In summary, a binary signaling scheme for a semiconductor radiation source has been described, in which the drive current is modulated by turning the drive current on and off, and thus also modulates the radiation output level. However, the modulation of the drive current is adapted to switch between two (or more) selected radiation output levels instead of waiting for the radiation response to follow a complete rise or fall curve. During an extended zero (or one) sequence, the drive current is switched to maintain the radiation output level between a minimum and maximum level. This range of radiation output levels is adaptively selected based on the characteristics of the drive current, the optical transmission channel, the radiation source, and / or the radiation detector.

[0152] While the invention has been detailed and described in the accompanying drawings and the foregoing description, such description should be considered illustrative or exemplary and not limiting. The invention is not limited to the disclosed embodiments. The proposed accelerometer keying concept can be applied to other types of optical wireless networks and used in conjunction with other types of access devices, modems, and transceivers. In particular, the invention is not limited to LiFi-related environments, such as ITU-T G.9961, ITU-T G.9960, and ITU-T G.9991 network environments. It can be used in visible light communication (VLC) systems, IR data transmission systems, G.VLC systems, OFDM-based systems, interconnected lighting systems, OWC systems, and smart lighting systems.

[0153] By studying the accompanying drawings, the disclosure, and the appended claims, those skilled in the art can understand and implement other variations of the disclosed embodiments in practicing the claimed invention. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite articles "a" or "an" do not exclude a plurality. A single processor or other unit can perform the functions of several items listed in the claims. The mere fact that certain measures are referenced in mutually different dependent claims does not imply that combinations of these measures cannot be used advantageously. The foregoing description has detailed certain embodiments of the invention. However, it will be appreciated that the invention can be practiced in many ways, however detailed it may appear in the text, and is therefore not limited to the disclosed embodiments. It should be noted that the use of particular terms when describing certain features or aspects of the invention should not be construed as implying that the term is redefined herein to be limited to include any specific characteristic of the feature or aspect of the invention associated with that term.

[0154] A single unit or device can perform the functions of several items listed in the claims. The mere fact that certain measures are referenced in mutually different dependent claims does not imply that a combination of these measures cannot be used advantageously.

[0155] Similar to Figure 3 The processes indicated herein can be implemented as program code devices of a computer program and / or dedicated hardware of a receiver or transceiver device. The computer program can be stored and / or distributed on a suitable medium, such as an optical storage medium or a solid-state medium, provided together with or as part of other hardware; however, it can also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems.

Claims

1. A transmitter (10) for generating radiated signals in an optical communication system, the transmitter comprising: A device for driving a semiconductor radiation source (12), the device comprising: a modulator (14) configured to: - A driving signal provided to the radiation source (12) is modulated with a data sequence to generate a driving current of at least two levels through the radiation source (12), the at least two levels of driving current defining a radiation output range between an upper radiation output level and a lower radiation output level of the radiation source (12) during the transmission of the data sequence; and The controller (18) is configured as follows: - The transmission quality of the radiation output at the transmitter is determined using at least one of a feedback signal indicating the radiation output level of the radiation source and feedback information received from the receiver of the radiation output. - Based on the transmission quality of the radiated output, determine the upper target level and lower target level that define the reduced radiated output range; and - Adaptively control the modulator (14) to control the drive current, and switch the drive current at a timing determined by a reduced radiation output range between a predetermined upper target level and a predetermined lower target level.

2. The transmitter of claim 1, wherein the controller (18) is configured to maintain the upper target level and the lower target level by switching the drive signal at a rate higher than the rate of the data sequence.

3. The transmitter of claim 2, wherein the data sequence is a binary data sequence, and wherein the controller (18) is configured to control the modulator (14) to apply a binary switching mode as the drive signal during the bit duration of the binary data sequence, and wherein the binary switching mode is determined by a combination of the current bit value of the binary data sequence and at least one previous bit value.

4. The transmitter of claim 1, wherein the transmission quality of the radiated output is based on the feedback signal, and the modulator (14) is configured to compare the feedback signal with an inversion lower limit and an inversion upper limit, and to turn off the drive signal when the feedback signal has reached the inversion upper limit and / or turn on the drive signal when the feedback signal has reached the inversion lower limit.

5. The transmitter according to claim 1, wherein the modulator (14) is configured to apply a short-circuit or negative drive current to the radiation source (12) during the off state of the drive signal.

6. The transmitter of claim 1, wherein the data sequence is a binary data sequence, and wherein the controller (18) is configured to control the modulator (14) to insert a sub-bit period or pattern of a logic value different from the current logic value of the binary data sequence into the drive signal to achieve the upper target level or the lower target level at the sampling time.

7. The transmitter according to claim 1, wherein the controller (18) is configured to select the upper target level and the lower target level such that the rise time of the radiation output from the lower radiation output level to the upper radiation output level is equal to the fall time of the radiation output from the upper radiation output level to the lower radiation output level.

8. The transmitter of claim 1, wherein the modulator (14) is configured to modulate the drive signal to produce more than two radiated output levels, wherein the transitions in the grid of the multi-level drive mode are adapted to the rising and falling curves of the radiated source.

9. An optical communication system comprising a transmitter (10) according to claim 1 and a receiver (20) for receiving a radiated signal generated by the transmitter (10).

10. The system of claim 9, wherein the receiver (20) is configured to detect the transmission quality of the radiated signal and feed back information about the detected transmission quality to the transmitter (10) to control the radiation output range of the radiation source (12).

11. A method for driving a semiconductor radiation source (12) at a transmitter used to generate a radiation signal in an optical communication system, the method comprising: The driving signal provided to the radiation source (12) is modulated by a data sequence to generate a driving current of at least two levels through the radiation source (12), the driving current of at least two levels defining the radiation output range between the upper radiation output level and the lower radiation output level of the radiation source (12) during the transmission of the data sequence. The transmission quality of the radiation output at the transmitter is determined using at least one of feedback information received from the receiver of the radiation output and a feedback signal indicating the radiation output level of the radiation source. The upper and lower target levels for limiting the reduced radiation output range are determined based on the transmission quality of the radiation output. and The drive current is adaptively controlled and switched at a timing determined by the reduced radiation output range between the upper target level and the lower target level.

12. A computer program product comprising code that, when executed by a controller (18) of a transmitter (10) according to claim 1, implements the steps of the method according to claim 11.

Images