What is crosstalk in wireless systems

EP1897252B1 - Dynamic data rate adaptive signal processing method in a wireless infrared data transmission system - Google Patents

Dynamic data rate adaptive signal processing in a wireless infrared data transmission system Download PDF

info

Publication number
EP1897252B1
EP1897252B1EP20060742293EP06742293AEP1897252B1EP 1897252 B1EP1897252 B1EP 1897252B1EP 20060742293 EP20060742293 EP 20060742293EP 06742293 AEP06742293 AEP 06742293AEP 1897252 B1EP1897252BEP1 18
Authority
EP
European Patent Office
Prior art keywords
sub
carrier
data signal
electrical
modulation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP20060742293
Other languages
English (en)
French (fr)
Other versions
EP1897252A1 (de
Inventor
Jelena Grubor
Volker Jungnickel
Klaus-Dieter Langer
Clemens Von Helmolt
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fraunhofer Society for the Demand of Applied Research eV
Original assignee
Fraunhofer Society for the Demand of Applied Research eV
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to DE200510030299priorityCriticalpatent / DE102005030299B4 / de
Application filed by Fraunhofer Society for the Demand of Applied Research eVfiledCriticalFraunhofer Society for the Demand of Applied Research eV
Priority to PCT / DE2006 / 000750 priority patent / WO2006136126A1 / de
Publication of EP1897252A1publicationCriticalpatent / EP1897252A1 / de
Application granted granted Critical
Publication of EP1897252B1publicationCriticalpatent / EP1897252B1 / de
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Left

  • 230000003044adaptiveEffects0.000titleclaimsabstractdescription40
  • 238000003672 processing methodMethods0.000titleabstractdescription15
  • 230000000051modifyingEffects0.000claimsabstractdescription103
  • 230000003287opticalEffects0.000claimsabstractdescription73
  • 239000000969carrierSubstances0.000claimsabstractdescription56
  • 230000001702transmitterEffects0.000claimsabstractdescription32
  • 238000004422calculation algorithmMethods0.000claimsdescription28
  • 238000000034methodsMethods0.000claimsdescription12
  • 238000011068loadMethods0.000abstractdescription26
  • 230000004301light adaptationEffects0.000abstractdescription9
  • 230000005540biological transmissionEffects0.000description103
  • 238000004891communicationMethods0.000description14
  • 238000010586diagramMethods0.000description12
  • 230000004044responseEffects0.000description9
  • 230000033590base-excision repairEffects0.000description6
  • 238000004364calculation methodMethods0.000description5
  • 238000005516 engineering processMethods0.000description5
  • 230000000875correspondingEffects0.000description4
  • 238000005457optimizationMethods0.000description4
  • 238000001514detection methodMethods0.000description3
  • 230000000694effectsEffects0.000description3
  • 230000003595spectralEffects0.000description3
  • 235000010357aspartameNutrition0.000description2
  • 230000001419dependentEffects0.000description2
  • 238000005259measurementMethods0.000description2
  • 230000003068staticEffects0.000description2
  • 210000003491SkinAnatomy0.000description1
  • 230000002411adverseEffects0.000description1
  • 238000004458 analytical methodMethods0.000description1
  • 238000006243 chemical reactionMethods0.000description1
  • 230000018109developmental processEffects0.000description1
  • 230000004069differentiationEffects0.000description1
  • 238000003379elimination reactionMethods0.000description1
  • 238000011156evaluationMethods0.000description1
  • 230000002452interceptiveEffects0.000description1
  • 239000011159matrix materialSubstances0.000description1
  • 238000011160researchMethods0.000description1
  • 238000004088simulationMethods0.000description1
  • 238000000844transformationMethods0.000description1
  • 230000001131 transformingEffects0.000description1

Images

Classifications

    • H-ELECTRICITY
    • H04 — ELECTRIC COMMUNICATION TECHNIQUE
    • H04B TRANSMISSION
    • H04B10 / 00 — Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10 / 11 — Arrangements specific to free-space transmission, i.e. transmission through air or vacuum
    • H04B10 / 114 — Indoor or close-range type systems
    • H04B10 / 1143 — Bidirectional transmission

Abstract

Description

  • The invention relates to a dynamic data rate adaptive signal processing method for the transmission of data signals with a predetermined signal power in a wireless infrared data transmission system with an omnidirectional, bidirectional optical transmission channel with infrared light of a predetermined frequency between a stationary transmitter and a mobile receiver in the interior area.
  • The advantages of infrared transmission - very large, free and available bandwidth, no interfering interference with radio systems, use of the same working frequencies in neighboring transmission systems, high data security, defined limitation of data transmission on walls - make it a useful and attractive technology for indoor communication, for example in offices , Hospitals or airplanes.
  • In the research area of ​​wireless, non-directional infrared communication in indoor areas, known systems achieve a high data rate through angular or spatial diversity. Optical systems with diversity known from the prior art, however, require relatively complex optical systems (angle-diverse receivers or multi-beam transmitters or both). Therefore, electronic signal processing methods that are used to improve radio systems are also useful for infrared transmission. However, since these two media have fundamentally different properties, a simple application of the known radio channel techniques to the infrared channel cannot take place, but requires separate investigations. Due to the high dynamics in terms of bandwidth and performance of the transmission response of the infrared channel, which primarily depends on the quality of the line of sight (LOS) between transmitter and receiver, it is difficult to achieve a good signal-to-noise ratio for high data rates Use of reasonable data signal powers. In order to ensure user use and user mobility without signal interruptions, the known infrared systems are statically designed in such a way that reliable signal propagation is achieved throughout the interior. The system is designed for the worst case of propagation. As a result, however, a large amount of the channel capacity, which is available under good transmission conditions, remains unused. However, in order to maximize the system data rate and still ensure reliable transmission and full space coverage, the transmission system must be designed to be adaptive with regard to the data transmission rate, that is, it must be dynamized. If the behavior of the transmission channel is known, this allows, under disadvantageous conditions, a reduction in the transmission speed until a given error probability is reached. With adaptive signal processing, the data is processed and sent in relation to time or frequency, based on a channel estimate. The current channel properties are fed back from the receiver to the transmitter according to the time or frequency reference via a return channel and included in the adaptive signal processing there. An adaptive system therefore represents a closed control loop and effects an automatic adjustment of the data transmission rate in the time or frequency domain depending on the current transmission quality in the channel. By designing it for the current channel status and not for the worst channel status, a significantly higher data rate and better utilization of the signal power can be achieved.
  • A physical model of the wireless infrared data transmission system with an omnidirectional optical transmission channel between a stationary transmitter and a mobile receiver in the indoor area is taken from the Publication I "A Physical Model of the Wireless Infrared Communication Channel" by V. Jungnickel et al. (in IEEE Journal on Selected Areas in Communications, Vol. 20, No.3, April 2002, p. 631-640). Measurements carried out for such a communication system were already in the Publication II "Indoor Propagation Measurements at Infrared Frequencies for Wireless Local Area Networks Applications" by H. Hashemi et al. (in Proc. of Vehicular Technology Conference, vol. 43, No. 3, August 1994, pp. 562-576). The use of multiple transmitters in combination with angle-diverse receivers to achieve angular diversity in an omnidirectional optical transmission channel is out of the Publication III "Angle Diversity for Nondirected Wireless Infrared Communication" by J.B. Carruthers et al. (in IEEE Transactions on Communications, Vol. 48, No. 6, June 2000, pp. 960-969). A method is disclosed with a static channel in the interior, which is designed for the worst case of transmission. The disadvantage here is, on the one hand, the unsatisfactory utilization of the available channel capacity and, on the other hand, the relatively large outlay on equipment due to a tracking system and complex optics. The desire for efficient utilization of the channel capacity leads to the design of an adaptive system. The achievement of directed connections between transmitter and receiver by tracking systems to achieve spatial diversity is from the Publication IV "Electronic Tracking for Wireless Infrared Communications" by V. Jungnickel et al. (in IEEE Transactions of Wireless Communications, Vol. 2, No. 5, Sept. 2003, pp. 989-999). The possibility of switching between two modes (direct and diffuse) is described here, whereby the channel capacity can be better used, but a complex electronic tracking and pointing system is still required. When system adaptivity is achieved exclusively through signal processing, it is possible to achieve high transmission speeds with efficient use of the channel capacity without complex electronics.
  • So far, in known infrared systems, a data rate adaptation in the time domain has been processed with a serial baseband transmission. A change in the transmission speed is made possible by adaptive coding using a variable code length or rate (cf. Publication V "Rate-Adaptive Modulation Techniques for Infrared Wireless Communication" by L. Diana et al. (in Proc. of IEEE Intl. Conf. on Communications, Vancouver, B.C., Canada, pp. 597-603, June 1999) and the Publication VI "Performance Evaluation of Rate-Adaptive Transmission Techniques for Optical Wireless Communication" by J.M. Garrido-Balsells et al. (in Proc. of Vehicular Technology Conference, pp. 914-918, 2004) or through a combination of adaptive coding and power-efficient modulation (degree of L-PPM) (cf. Publication VII "Rate-Adaptive Indoor Infrared Wireless Communications Systems Using Punctured Convolutional Codes and Adaptive PPM" by M. Matsuo et al., 1998. The Ninth IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, Volume 2, 8-11 Sept. 1998 , pp. 693 - 697 vol. 2). However, such wide baseband transmission is very susceptible to inter-symbol interference because of the multipath in an infrared system. In addition, adaptive adjustment in the form of complex signal processing at the receiver is required. Furthermore, systems are proposed which deal with a combination of adaptive serial transmission with angle diversity (cf. Publication VIII "Angle diversity and rate adaptive transmission for indoor wireless optical communications" by A. Tavares et al., Communications Magazine, IEEE Volume 41, Issue 3, March 2003, pp. 64-73). However, this again requires complex optics to generate diversity and adaptive adjustment for serial transmission. Applications for infrared transmission systems for data rate adaptive transmission in the time domain are those mentioned above Publications V and VI to be taken from which the present invention is based as the closest prior art. However, the adaptive adjustment of the data rate takes place exclusively in the time domain by adapting the transmission speed of the serial data, determining it in blocks with a change in the block speed, the variation in the code word length or the complex coding to change the data rate.
  • In wireless communication systems based on radio, adaptive signal processing with the advantages of adaptation to the propagation conditions and higher transmission capacities both in the time domain and in the frequency domain is known. The required return channel is also constantly redefined depending on the current channel quality. Radio systems cannot be compared with optical systems simply by analogy, since the channels show fundamentally different transmission behavior (radio channel: modulation by changing the field strength (signal amplitude), positive and negative values ​​in complex are possible, optical channel: only positive values ​​in real terms are possible , direct modulation only by changing the light intensity, ie the amplitude of the optical power. The distinction between adaptivity in the time and frequency domains relates to whether the user has the data serially on one frequency (time approach) or split in parallel on different frequencies ( In the case of radio channels, these two methods are called single carrier and multiple carrier transmission technology. In an optical infrared system that works with one wavelength, however, there is only one optical carrier. This optical carrier can be directly connected to serial The data are modulated (baseband transmission, corresponding to the single carrier technology in the radio channel). Another method is to first split the electrical data stream in order to modulate a large number of electrical carriers in parallel. The optical carrier can then be modulated with this entire modulated electrical signal (multi-carrier modulation, corresponding to multi-carrier modulation in the radio channel).
  • A detailed description of the adaptive signal processing in the frequency domain for a radio channel is given in Publication IX: "Adaptive Multicarrier Modulation: A Convenient Framework for Time-frequency Processing in Wireless Communications" by Th. Keller et al. (in IEEE Proc. of the IEEE, Vol.88, No.5, May 2000, pp. 611-640). The algorithms used in the cited publications make it clear that a simple analogy between radio channel and optical channel is not possible.
  • The adaptive signal processing with multiple carriers is called DMT (Discrete Multi-Tone) for the wired electrical transmission channel, AOFDM (Adaptive Orthogonal Frequency Division Multiplexing) for the wired radio channel and MOSM-IR (Multiple Orthogonal Subcarrier Modulation) for the optical transmission channel, an additional function block for intensity modulation or direct detection is required here on both the transmit and receive sides. In the case of wire-free systems, it is fundamentally a matter of minimizing signal distortion caused by multipath propagation. This involves the same techniques (multiple carriers, adaptive selection of the modulation format for each carrier, return channel for the control information), with DMT crosstalk and AOFDM supposed to suppress multipath propagation. In both techniques, multiple carriers are used to solve different physical problems that occur in the physically different transmission channels.
  • The object of the present invention is therefore to develop the generic signal processing method explained above in such a way that adaptive signal processing in a wireless infrared data transmission system, in which the properties of the optical transmission channel are known to fluctuate in a large frequency range, are also carried out in the frequency range can, in order to be able to achieve the advantages of the radio channels through adaptation in the frequency range even with optical transmission channels. The solution according to the invention for this problem can be found in the main claim. Advantageous further developments are shown in the subclaims and explained in more detail below in connection with the invention.
  • In the generic dynamic data rate adaptive signal processing method for the transmission of data signals with a predetermined signal power in a wireless infrared data transmission system with an omnidirectional, bidirectional optical transmission channel of infrared light of a predetermined frequency between a stationary transmitter and a mobile receiver in the interior area is provided that
    • the electrical data signal is divided into a number N parallel data signal streams and assigned to a large number of orthogonal electrical subcarriers of different subcarrier frequencies,
    • the data rate of the data signal streams currently to be transmitted for each electrical subcarrier is determined quasi-continuously by individually selecting a modulation format and the data transmission speed depending on the respective transmission quality of the subcarrier, with the optical signal power being distributed to the individual subcarriers in compliance with safety regulations
    • The intensity of the infrared light to be transmitted is modulated directly according to the selected modulation formats after parallel-to-series conversion,
    • the serially transmitted optical signal is detected directly in the baseband at the receiver,
      the number N of parallel electrical data signal streams is demodulated,
    • each electrical data signal stream is processed separately and that
    • the subcarrier with the lowest subcarrier frequency is used permanently as a return channel from the receiver to the transmitter for the transmission of control information regarding the current transmission quality of each subcarrier for the adaptive determination of the current data rate to be transmitted in each subcarrier.
  • The data rate adaptive signal processing method according to the invention with a dynamic determination of the current transmission rate per subcarrier enables data processing in the frequency range for the first time for transmission optimization in a simple optical transmission system without complex optical or electronic components. The improvement of the transmission, which is possible bidirectionally in the transmission channel, is achieved exclusively through signal processing. Orthogonal electrical subcarriers are multiplexed in the frequency range. The modulation and, accordingly, the demodulation of the data signals takes place in two stages. At the transmitter, an electrical and then an optical modulation is carried out first. On the receiver side, the demodulation takes place in reverse order. Furthermore, an additional adaptive coding of the subcarriers can take place if an uncoded transmission is considered. The transmission capacity made available by the channel is used efficiently: under good transmission conditions, very high data rates can be transmitted; under poor transmission conditions, on the other hand, the data rates are reduced. The result is a benefit-cost-optimized optical transmission channel with high quality of service (QoS). In the method according to the invention, the return channel is permanently placed on the electrical subcarrier with the lowest subcarrier frequency that ensures transmission. This assignment can reliably ensure permanent feedback in the system, which can also be easily checked. If the return channel is deliberately interrupted, the failure of the adaptation due to an abrupt deterioration in the signal transmission and thus the effectiveness of the adaptation system can be recognized immediately. An analyzer in the return channel can detect how the control information changes when the transmission channel changes.
  • The aim of the method according to the invention was to solve the problem of solving the multipath propagation in a wireless connection with an infrared channel (with positive real values ​​of the light intensity), whereby the conditions for the optical signal to be transmitted are tightened compared to known transmission methods. By dividing it into N parallel subcarriers, the transmission speed per subcarrier can be reduced, which is equivalent to a lower bandwidth. As a result, the influence of the inter-symbol interference can be reduced, which in turn has a positive effect on the receiver, on which only simple signal processing has to be carried out. Further details on the method according to the invention can be found in the special part of the description. There are also explanations of embodiments of the signal processing method that deal with the specific selection of the modulation formats for each subcarrier according to a bit loading algorithm under real operating conditions (instantaneous channel noise, maximum permissible optical signal power, permissible BER).
  • The mode of operation of the data rate adaptive signal processing method according to the invention is explained in more detail below with reference to the schematic figures. It shows:
  • Figure 1
    a representation of a small office with an infrared data transmission system,
    Figure 2
    a diagram of the frequency-dependent infrared channel response with different K-factors,
    Figure 3
    a diagram of the K-factor distribution for different ranges and recipient orientations,
    Figure 4
    a block diagram of the procedure,
    Figure 5
    a diagram for the optimal power distribution for different channel states,
    Figure 6
    a diagram of the total maximum possible data rate depending on the K-factor,
    Figure 7
    a BER diagram for different modulation formats,
    Figure 8
    a diagram of the channel capacity as a function of the channel condition,
    Figure 9A
    an OPC diagram for an IR channel,
    Figure 9B
    an OPC diagram for a DSL channel,
    Figure 10
    an example of different bit distributions for the two-stage bit loading algorithm and
    Figure 11
    a flow chart for an optimal bit loading algorithm.
  • A typical application of an infrared wireless transmission system is shown in FIG. 4 in the form of a small office. The transmitter Tx is arranged in the middle of the ceiling, the receiver Rx can be arranged as desired, for example on a desk, and has any orientation while maintaining the line of sight LOS to the transmitter Tx. Such an arrangement with an undirected transmission channel advantageously enables both a diffuse and a direct communication link between transmitter Tx and receiver Rx. If there is a direct connection between them, line of sight can be used for communication. If the direct connection is broken, the diffuse connection can maintain communication at a reduced data rate.
  • The transfer function of the infrared channel in such an undirected line of sight results from:
    with ηCOME ON = Proportion of the line of sight with regard to path amplitude attenuation, which is fundamentally independent of the modulation frequency ƒ is and Hdiff (ƒ) = Part of the diffuse reflection in the room, which can be estimated using a first order low-pass function. It can be seen that the channel response has great dynamics in the frequency domain, which depends on the power ratio between the line of sight and the diffuse signal. This dependency can be combined with the K-factor in dB K = 20.log (ηCOME ON/ ηdiff) can be quantified, where ηdiff the distribution of the diffuse connection with respect to path amplitude damping st.
  • The channel dynamics with the parameters frequency and K-factor for the office room are shown in FIG. The diagram shows the course of the logarithmically scaled and standardized channel amplitude over the transmission frequency for various K-factors. If the line of sight is blocked or very poor, the channel response corresponds approximately to a low-pass filter and the channel bandwidth is low. As the line of sight improves, the channel response varies greatly and shows cuts until it runs smoothly for sufficient K-factors and has a bandwidth that is an order of magnitude larger than in the diffuse case.
  • The scenario according to is selected in such a way that the values ​​of the K-factors are obtained for real ambient conditions. The distribution of the K-factors in space is shown in FIG. 3 for three different values ​​of the Lambert radiation index m at the transmitter Tx, which determine the beam width and limit the area of ​​the line of sight if this is not blocked. The shows the course of the K-factor values ​​as a function of the horizontal distance r between transmitter Tx and receiver Rx and the directional effect Ψ at the receiver angle. It can be seen that the K-factor range is larger with a better alignment of the transmitter Tx (larger m), but the range for a possible line of sight is reduced, so that several transmitters would be required to safely cover the entire room. It can also be seen from this, however, that for m = 1 for a transmission channel with moderate directivity, there is poor transmission only in the corners of the room. Nevertheless, higher data rates can be achieved with the method according to the invention than for a diffuse channel. The increase in the K-factor takes place more directly for a more strongly directed transmitter Tx. At the same time, the directional angle to the receiver Rx is more critical than its distance to the transmitter Tx.
  • The basic block diagram for the signal processing method according to the invention can be found in FIG. According to a communication between a transmitter fixed on the room ceiling Tx and a mobile receiver Rx accepted. The electrical data signal data will be on the transmitter Tx in N divided parallel currents, which are then divided into a corresponding number N be modulated on by subcarriers in a suitable manner. The optimal modulation format for each subcarrier is determined in accordance with control information in the form of a vector of the modulation codes per subcarrier via a return channel RC at low speed in the uplink from the mobile receiver Rx to the stationary transmitter Tx be sent, selected. Various modulation methods for optical signal transmission, for example BPSK, QPSK, OOK, LPPM, are known from the prior art. After the adaptive initial modulation APM the data symbols to be transmitted are made by digital signal processing DMTM generated. DMT modulation is well known in the art for conducted electrical transmission channels. In the case of the optical transmission channel, DMT is used to directly modulate a light source LD used. Before that, the processed signals are given a DC component in a further functional block DC added to the light source LD to be controlled correctly.
  • The modulated optical signals then reach the mobile receiver Rx, after going from the optical transmission channel IRWC and the natural ambient light that causes noise. At the recipient Rx there is first a direct signal detection by means of a photodiode PD, then the constant component DC eliminated, DCB, and a DMT demodulation, DMTD, carried out. Continue to all N Subcarriers processed individually. A part of the signal power in each subcarrier is subtracted for the channel estimation CE. Then the data is balanced, EQ, which can be done by multiplying by the inverse channel coefficients. Different ways of channel estimation and signal adjustment are generally sufficiently known from the prior art. Furthermore, the recipient Rx the current vector for the best modulation formats per subcarrier is generated depending on the current channel status and via the return channel RC to the transmitter Tx cleverly. The vector can be generated as follows: the data adjustment EQ yields a vector for the signal-to-noise ratio SNR or noise amplification, on the amount of which a suitable bit loading scheme is implemented to determine the number of bits and the distribution of the signal power on each subcarrier with reference to the constellation conditions, the modulation format and set the alphabet size to determine. Various bit loading methods are also known from the prior art. For a good subcarrier that has good channel attenuation and a relatively low increase in noise, a modulation scheme with greater bandwidth efficiency is used, so that a large amount of information can be transmitted. On the other hand, a bad subcarrier will transmit little or no information at all. However, transmission with a predetermined bit error rate can be implemented despite the channel weakening. The return channel RC can occupy the lowest subcarrier that is still suitable for transmission in order to realize the necessary feedback for the implementation of the adaptive system.
  • The capacity analysis of the subcarriers for optimal power distribution is carried out as follows. It follows directly from the fact that intensity modulation and direct detection are used in an infrared system that the electrical signal-to-noise ratio SNR is proportional to the square of the mean value of the received optical power. The electrical signal-to-noise ratio SNR is defined as the ratio of the electrical signal power P.cl, R and the electrical noise power Ncl at the receiver Rx, which results in:
    in which I.ph the generated photocurrent, σ2 the noise variance, R. the receiver sensitivity and |H| is the amplitude of the channel frequency response, while the optical power denotes Rx at the receiver and Tx at the transmitter. It is believed that the electrical noise power comes from backlight noise as the most significant contributor in the infrared system (at 900 nm).
  • An infrared system with N orthogonal subcarriers can be viewed as a system of N parallel independent Gaussian channels with independent noise. The total system capacity (bits / transmission), according to the formulas of Shannon's information theory, results from the sum of the capacities of all subcarriers, which, after inserting equation (2), follows:
    in which P.opt, T,i , |Hi| and αi = α2 · R.-2·|Hi|-2 represent the optical power, the amplitude of the channel response and the effective noise of the i-th subcarrier. It is important to recognize that the term of the optical power according to equation (3) is squared. This results in a serious difference in the capacity function between the optical channel and the radio channel in which the corresponding term is not squared. This results in the need for a special power distribution scheme, which is described below.
  • Infrared transmission systems are dependent on safety regulations (eye and skin protection), which set an upper limit for the average total power to be transmitted by the transmitter Tx P.DEAD establish. This overall performance P.DEAD must now be distributed to the independent subcarriers in such a way that a maximum system capacity is achieved. This is an optimization problem known from the prior art that can be solved with the aid of Lagrange factors.
  • In the mathematical calculation, a vector P.=(P.1, P.2, ..., P.N) can be determined so that the N-variable function C.(P.) is maximized according to equation (3) under the condition First, the Lagrange function is formed to L.(P.) = C.(P.) -λG(P.), where λ = const.∈ applies to the Lagrange factor R. . The optimized solution can then be derived from a system of N equations ∂L (P.i)/∂P.i =0, i =1... N can be obtained, which after some transformations result in:
    with "otherwise" meaning for all other cases outside of the case mentioned in the first line.
  • The optimal power distribution for three different channel states (different K-factors) is shown in. In a simulation, 128 subcarriers were distributed over a bandwidth of 300 MHz, the Si-APD (with a factor M = 100 and a highly assumed noise factor x = 0.35) with an effective total surface area AR.= 30 mm2 accepted. The optical performance condition is set in accordance with the safety regulations for extended diffuse light sources at 900 nm P.DEAD = P.Tx = 400 mW. For an almost diffuse (with a low K-factor) or diffuse transmission channel, it can be seen that the signal power is only distributed over the few first subcarriers. If the channel behavior improves towards a larger bandwidth, more and more subcarriers are filled up with a lower power and used for the transmission. For K-factors with a broad channel response, the result is an even distribution of power over all subcarriers.
  • The potential data rates of an infrared transmission system with the adaptive signal processing described above for three different values ​​of the transmission power P.Tx are shown in the.As a comparison, graphs are shown for a static subcarrier system in which transmission is reliably guaranteed in the entire room with a diffuse transmission channel. The advantages of adaptive signal processing can be clearly seen when the K-factors reach the 0 dB range and increase in such a way that a large increase in the data rate can be achieved for moderate values ​​of the K-factor. After reaching the K-factors of 15 dB, the advantages increase even more. For greater powers of the transmitter Tx, the difference between adaptive and non-adaptive signal processing is particularly noticeable. It should be noted that with a scaling according to, an optimal distribution of the potential data rates in a room can also be obtained according to the results according to.
  • Such an optimization is known for maximizing capacity in wireless radio channels and is referred to as the "water filling principle". The capacity is maximized in accordance with
    under the electrical power condition, which differs fundamentally from the capacity function to be maximized in the infrared channel and is linearly dependent on the power. Since the capacity of an infrared channel, just like a radio channel, can be expressed as a function of the electrical signal power at the transmitter Tx, it makes sense to try to achieve an optimal power distribution according to the water-filling principle. However, it could be shown that the optimal power distribution has a different appearance and that the water-filling principle in infrared transmission, which is limited in optical power by the safety regulations, only leads to suboptimal power distributions. In the case of a transmission power below the safety conditions, for example due to the hardware used, the water-filling principle leads to a maximum capacity utilization of the system, even with infrared systems, but not to a maximum data rate. The optical conditions in an infrared system must therefore be carefully considered. The system parameters can then be optimized with regard to the number of subcarriers, the choice of suitable bit loading systems and modulation formats, the sufficient level of modulation, degree of adaptation, etc. to achieve the theoretical maximum capacity in an infrared transmission system .